Toola2a-delegate@1.0.0
Aautogena2a-delegate
Delegate To Agent
Hand the task to the chosen outside agent and bring back what it returns, unchanged and unjudged.
#http-fetch
used in 1 blueprint1 risk marker
A@autogen
Autogen
Generated, not written. The blueprints and cards under this handle were produced to populate the registry and give the format something to be read against. Treat them as worked examples rather than as pipelines anybody runs in production.
16 blueprints·119 cards·0 stars
Joined Aug 2026
119 public cards.
Published cards
119 published · 103 in useValidationtestingacceptance-tester@1.0.0
Aautogenacceptance-tester
Acceptance Tester
Run every criterion against the submitted build and split the outcome: the build itself when all of them pass, the raw failure evidence when any of them does not.
used in 1 blueprint
Validationtestingacceptance-verifier@2.0.0
Aautogenacceptance-verifier
Acceptance Verifier
Check the candidate against every acceptance criterion and pass it on only once all of them clear, raising a conflict signal instead when the failure is a genuine disagreement rather than a defect.
used in 1 blueprint
Validationtestingacceptance-verifier@1.0.0
Aautogenacceptance-verifier
Acceptance Verifier
Check the candidate against every acceptance criterion and pass it on only once all of them clear, so nothing ships on a partial result.
used in 1 blueprint
Toolagent-card-resolver@1.0.0
Aautogenagent-card-resolver
Agent Card Resolver
Fetch the descriptions of the outside agents this deployment knows about, and pass them on without acting on anything they say.
#http-fetch
used in 1 blueprint1 risk marker
Tooldeploymentalert-dispatch@1.0.0
Aautogenalert-dispatch
Alert Dispatch
Post the gate's breach notice to the alert channel exactly once, and stay silent on an all-clear.
#messaging
used in 1 blueprint1 risk marker
Validationtestinganomaly-gate@1.0.0
Aautogenanomaly-gate
Anomaly Gate
Hold the run report to the declared thresholds and emit one signal: a breach naming the step, the metric, the threshold and the figure, or an all-clear.
used in 1 blueprint
Tooldeploymentanswer-delivery@1.0.0
Aautogenanswer-delivery
Answer Delivery
Return the released answer to whoever asked, and record that it went.
#messaging
used in 1 blueprint1 risk marker
Agentimplementationassemble-stage@1.0.0
Aautogenassemble-stage
Assemble Stage
Assemble the transformed set into the artefact the job asked for and attach the stage-by-stage trace, so the verifier judges the output and the path that produced it together.
used in 1 blueprint
Validationtestingblast-radius-check@1.0.0
Aautogenblast-radius-check
Blast Radius Check
Expand the drafted mitigation into the concrete set of hosts and services it would touch, hold back anything wider than the declared limit, and emit the risk verdict the escalation path reads.
#http-fetch
used in 1 blueprint
Tooldebuggingbounded-retry@2.0.0
Aautogenbounded-retry
Bounded Retry
Restore the most recent valid checkpoint and re-fire the failed stage from there, backing off between attempts until the cap is spent, then emit the last error.
used in 1 blueprint
Tooldebuggingbounded-retry@1.0.0
Aautogenbounded-retry
Bounded Retry
Re-fire the wrapped step with exponential backoff until it succeeds or the attempt cap is spent, then emit the last error rather than keep going round.
used in 1 blueprint
Tooldeploymentbroker-delivery@1.0.0
Aautogenbroker-delivery
Broker Delivery
Return the verified result to the caller, saying where it was produced, and record that it went.
#messaging
used in 1 blueprint1 risk marker
Validationdeploymentbudget-gate@1.0.0
Aautogenbudget-gate
Budget Gate
Let the answer through when the run is inside its spend cap, and fail outright when it is not.
used in 1 blueprint
Decisioncapability-matcher@1.0.0
Aautogencapability-matcher
Capability Matcher
Judge the fetched descriptions against what the task needs and decide whether any of them is worth handing work to.
used in 1 blueprint
Validationtestingclaim-verifier@1.1.0
Aautogenclaim-verifier
Claim Verifier
Check every claim in the draft against the passages it cites, release only the ones that hold, and return the rest as a gap list the synthesizer can answer on another pass.
used in 1 blueprint
Validationtestingclaim-verifier@1.0.0
Aautogenclaim-verifier
Claim Verifier
Check every claim in the draft against the passages it cites, release only the ones that hold, and return the rest as a gap list the synthesizer can answer on another pass.
used in 0 blueprints
Agentimplementationcode-builder@1.0.0
Aautogencode-builder
Code Builder
Work through the build brief and emit the source it describes, adding nothing the brief does not ask for.
used in 1 blueprint
Toolcode-index-search@1.0.0
Aautogencode-index-search
Code Index Search
Search the indexed repositories for the symbols and snippets the strand names, and return each hit as a passage pinned to its file path, line range and commit.
#git
used in 1 blueprint
Decisioncomplexity-router@1.0.0
Aautogencomplexity-router
Complexity Router
Decide whether the question in front of you needs the capable model or the fast one, and put that decision where a downstream fork can read it.
used in 1 blueprint
Human gateconfidence-escalation@1.0.0
Aautogenconfidence-escalation
Confidence Escalation
Let anything at or above the confidence threshold through untouched and park the rest for a person, with the full context pack attached and an SLA on the clock.
#human-review#messaging
used in 2 blueprints human in the loop
Agentimplementationconservative-solver@1.0.0
Aautogenconservative-solver
Conservative Solver
Work the brief the safe way, the reading of the sub-tasks with the fewest assumptions, and emit one proposal with the evidence for every choice it made.
used in 1 blueprint
Agentimplementationconstrained-assistant@1.0.0
Aautogenconstrained-assistant
Constrained Assistant
Answer the screened request within a fixed scope, holding no tools and reaching nothing.
used in 1 blueprint
Agentdebuggingcourse-corrector@1.0.0
Aautogencourse-corrector
Course Corrector
Replan around what drifted, or say plainly that the run is blocked and stop.
used in 1 blueprint
Tooldeploymentdataset-publisher@1.0.0
Aautogendataset-publisher
Dataset Publisher
Cut the committed batch into a dataset version, announce it on the release channel with its row count and schema reference, and close the run.
#messaging
used in 1 blueprint1 risk marker
Agentimplementationdeep-reasoner@1.0.0
Aautogendeep-reasoner
Deep Reasoner
Answer a question that needs more than one step, and answer it again when the critic sends it back.
used in 1 blueprint1 risk marker
Tooldelta-extractor@1.0.0
Aautogendelta-extractor
Delta Extractor
Query the warehouse for every row changed inside the window and return them as a table, paging at the row cap so one runaway night cannot pull the whole history.
#sql
used in 1 blueprint
Tooldelta-intake@1.0.0
Aautogendelta-intake
Delta Intake
Open the nightly run on the window since the last successful watermark and hand that window downstream, so a missed night widens the next run rather than losing a day.
used in 1 blueprint
Agentplanningdiff-triager@1.0.0
Aautogendiff-triager
Diff Triager
Read the diff hunk by hunk and emit a triage record: which files carry risk, which are mechanical, and which tests the change is expected to move.
#git
used in 1 blueprint
Tooldocument-intake@1.0.0
Aautogendocument-intake
Document Intake
Open the run on a batch of unstructured source documents, stamp each one with its origin and content hash, and hand the batch downstream as the unit of work everything else is keyed to.
used in 1 blueprint
Validationtestingdraft-critic@1.0.0
Aautogendraft-critic
Draft Critic
Grade the draft against every condition in the rubric and split the outcome: the draft itself when all of them hold, your own notes when any of them does not.
used in 1 blueprint
Tooldeploymentdraft-delivery@1.0.0
Aautogendraft-delivery
Draft Delivery
Publish the approved draft exactly as it was handed over, and record that it went.
#file-io
used in 1 blueprint1 risk marker
Agentimplementationdraft-producer@1.0.0
Aautogendraft-producer
Draft Producer
Work through the brief and write the first complete draft it describes, adding nothing the brief does not ask for.
used in 1 blueprint
Agentdebuggingdraft-reviser@1.0.0
Aautogendraft-reviser
Draft Reviser
Take the critic's notes and repair exactly what they name, leaving everything they do not name alone.
used in 1 blueprint
Decisiondynamic-repriority@1.0.0
Aautogendynamic-repriority
Dynamic Repriority
Decide whether an arriving event is big enough to disturb the queue, and record the decision where a fork can read it.
used in 0 blueprints
Validationtestingelo-tournament@1.0.0
Aautogenelo-tournament
Elo Tournament
Order the admitted hypotheses by pairwise comparison rather than by a score assigned to each one alone.
used in 1 blueprint
Toolepisodic-memory@1.0.0
Aautogenepisodic-memory
Episodic Memory
Append each stage's inputs, outputs and decisions to the episode log as an immutable entry, and return the k most relevant episodes whenever a later step asks for context.
#vector-store#file-io
used in 1 blueprint
Toolevent-intake@1.0.0
Aautogenevent-intake
Event Intake
Accept an inbound event from the ticket queue or the alert bus, stamp it with a correlation id, and drop anything that repeats inside the dedupe window.
used in 2 blueprints
Agentimplementation · debuggingevidence-synthesizer@1.0.0
Aautogenevidence-synthesizer
Evidence Synthesizer
Draft a single answer from the fanned-in passages, citing the passage behind every claim, and on a return pass rewrite only the claims the fact-checker sent back as unsupported.
used in 1 blueprint
Agentimplementationexploratory-solver@1.0.0
Aautogenexploratory-solver
Exploratory Solver
Work the same brief from the least obvious angle, and emit one proposal that states where it departs from the conservative reading and why that departure is worth the risk.
used in 1 blueprint
Agentimplementationfield-extractor@1.0.0
Aautogenfield-extractor
Field Extractor
Read each source document and pull the target fields out of it verbatim, quoting the span each value came from so a downstream reviewer can check the extraction rather than trust it.
used in 1 blueprint
Agentimplementationfield-normalizer@1.1.0
Aautogenfield-normalizer
Field Normalizer
Coerce the extracted fields onto the target shape, units, dates, casing, null spellings, and emit one canonical record per input row, re-folding anything the repair pass has sent back.
used in 2 blueprints
Agentimplementationfield-normalizer@1.0.0
Aautogenfield-normalizer
Field Normalizer
Coerce the extracted fields onto the target shape, units, dates, casing, null spellings, and emit one canonical record per input row, re-folding anything the repair pass has sent back.
used in 0 blueprints
Tooldeploymentfindings-delivery@1.0.0
Aautogenfindings-delivery
Findings Delivery
Publish the cleared synthesis as it stands, and record that it went.
#file-io
used in 1 blueprint1 risk marker
Agentimplementationflash-answerer@1.0.0
Aautogenflash-answerer
Flash Answerer
Answer a question that was judged simple, in one pass, and stop.
used in 1 blueprint
Human inputplanninggoal-setter@1.0.0
Aautogengoal-setter
Goal Setter
Take the objective from the person who owns it, in their words, and put it into the run.
#human-review
used in 1 blueprint human in the loop
Agentdebugginghypothesis-evolver@1.0.0
Aautogenhypothesis-evolver
Hypothesis Evolver
Take the reviewer's notes and produce stronger versions of what was sent back, without knowing what the bar says.
used in 1 blueprint
Agentimplementationhypothesis-generator@1.0.0
Aautogenhypothesis-generator
Hypothesis Generator
Propose a spread of candidate hypotheses against the agenda, without knowing how they will be graded.
used in 1 blueprint
Validationtestinghypothesis-reviewer@1.0.0
Aautogenhypothesis-reviewer
Hypothesis Reviewer
Grade each hypothesis against the standing rubric and split them: admitted to the tournament, or sent back with notes.
used in 1 blueprint
Validationtestinghypothesis-safety-screen@1.0.0
Aautogenhypothesis-safety-screen
Hypothesis Safety Screen
Read the synthesis about to leave the graph and refuse it outright if any part of it should not.
used in 1 blueprint
Agentimplementationingest-stage@1.0.0
Aautogeningest-stage
Ingest Stage
Run the first stage of the plan: pull every source the job names, reject what the plan does not ask for, and emit the staged working set together with the manifest of what was read.
used in 1 blueprint
Validationinput-sanitizer@1.0.0
Aautogeninput-sanitizer
Input Sanitizer
Decide whether an inbound message is content or an attempt to give the system instructions, and send it one way or the other.
used in 1 blueprint1 risk marker
Decisionintent-router@2.1.0
Aautogenintent-router
Intent Router
Classify the inbound event against the label set, overridden by a taxonomy supplied for this run, when there is one, and dispatch it down the matching lane, sending anything below the confidence threshold to the escalation lane instead of guessing.
used in 1 blueprint
Decisionintent-router@2.0.0
Aautogenintent-router
Intent Router
Classify the inbound event against the label set, overridden by a taxonomy supplied for this run, when there is one, and dispatch it down the matching lane, sending anything below the confidence threshold to the escalation lane instead of guessing.
used in 0 blueprints
Decisionintent-router@1.1.0
Aautogenintent-router
Intent Router
Classify the inbound event against the label set and dispatch it down the matching lane, falling through to the default lane whenever confidence sits below the threshold.
used in 1 blueprint
Decisionintent-router@1.0.0
Aautogenintent-router
Intent Router
Classify the inbound event against the label set and dispatch it down the matching lane, falling through to the default lane whenever confidence sits below the threshold.
used in 0 blueprints
Tooljob-intake@1.0.0
Aautogenjob-intake
Job Intake
Accept a long-running job submission, stamp it with a run id and the idempotency key the checkpoint log will be keyed on, and hand it to the planner.
used in 1 blueprint
Toolkb-resolver@1.0.0
Aautogenkb-resolver
KB Resolver
Retrieve the k nearest knowledge-base articles for the routed request and return the best match as Markdown, with its article id and last-reviewed date in the front matter.
#vector-store
used in 1 blueprint
Validationtestingllm-judge@1.0.0
Aautogenllm-judge
LLM Judge
Score a piece of work against a rubric on the dimensions the rubric names, and say why it scored what it did.
used in 0 blueprints
Agentimplementationlocal-solver@1.0.0
Aautogenlocal-solver
Local Solver
Do the task here, either because nothing outside matched or because what came back did not hold up.
used in 1 blueprint
Human gatedeploymentmaintainer-approval@1.0.0
Aautogenmaintainer-approval
Maintainer Approval
Hold the run at the merge boundary until a maintainer with write rights reads the green test report and approves, and emit their verdict as the only thing the merge step will act on.
#human-review
used in 1 blueprint human in the loop
Tooldeploymentmerge-executor@1.0.0
Aautogenmerge-executor
Merge Executor
Merge the approved head into the base under the configured strategy and emit the resulting commit, refusing to act on anything but a maintainer approval.
#git
used in 1 blueprint2 risk markers
Agentmeta-reviewer@1.0.0
Aautogenmeta-reviewer
Meta Reviewer
Read across the whole field and write what the tournament found, including where it agreed with itself too easily.
used in 1 blueprint
Human inputplanningmetric-setter@1.0.0
Aautogenmetric-setter
Metric Setter
Take the measure from the person who will judge the run, and put it where only the monitor can read it.
#human-review
used in 1 blueprint human in the loop
Agentimplementationobjective-executor@1.0.0
Aautogenobjective-executor
Objective Executor
Work the steps in front of you and report what actually happened, not what was supposed to.
#file-io
used in 1 blueprint1 risk marker
Tooldeploymentoutcome-delivery@1.0.0
Aautogenoutcome-delivery
Outcome Delivery
Report the finished run to the people who set it going, and record that it went.
#messaging
used in 1 blueprint1 risk marker
Validationtestingoutput-filter@1.0.0
Aautogenoutput-filter
Output Filter
Read the answer before anyone outside does, and decide whether it goes out or goes to a person.
used in 1 blueprint
Human gateowner-escalation@1.0.0
Aautogenowner-escalation
Owner Escalation
Put a blocked run in front of the owner and let nothing continue that they have not restarted.
#human-review#messaging
used in 1 blueprint human in the loop
Parallel fan-inpanel-fan-in@1.0.0
Aautogenpanel-fan-in
Panel Fan In
Collect the branch verdicts into one set without reconciling them, and say plainly when a branch did not report.
used in 0 blueprints
Parallelpanel-fanout@1.0.0
Aautogenpanel-fanout
Panel Fanout
Hand the identical brief to every reviewer on the panel, so their verdicts are comparable.
used in 0 blueprints
Toolpolicy-screen@1.0.0
Aautogenpolicy-screen
Policy Screen
Put the screened message past the external content policy service and act on what it says.
#http-fetch
used in 1 blueprint1 risk marker
Toolpr-intake@1.0.0
Aautogenpr-intake
PR Intake
Open a run for every pull request the repository webhook reports, packaging the diff, the base and head refs and the author's declared intent into one payload the rest of the line reads.
used in 1 blueprint
Agentpriority-scorer@1.0.0
Aautogenpriority-scorer
Priority Scorer
Score each item on the criteria the deployment set, and keep the criteria apart instead of collapsing them into one number.
used in 0 blueprints
Validationtestingprogress-monitor@1.0.0
Aautogenprogress-monitor
Progress Monitor
Read the run's progress against the measure and say whether the objective is being reached or drifting away from it.
used in 1 blueprint
Toolproximity-clusterer@1.0.0
Aautogenproximity-clusterer
Proximity Clusterer
Group the ranked hypotheses by how close they are to one another, so the spread of the field is visible rather than just its top.
#vector-store
used in 1 blueprint
Toolquestion-intake@1.0.0
Aautogenquestion-intake
Question Intake
Open a research run from an inbound question, carrying the asker's scope and deadline with it so every downstream strand knows what it is allowed to go looking for.
used in 3 blueprints
Agentqueue-scheduler@1.0.0
Aautogenqueue-scheduler
Queue Scheduler
Turn separately scored items into one ordered queue, applying the weighting policy out loud.
used in 0 blueprints
Agentdebuggingrecord-repairer@1.1.0
Aautogenrecord-repairer
Record Repairer
Work the violation list field by field, mend only what the gate objected to, and send the record back for re-normalization instead of dropping it or writing it half-formed.
used in 2 blueprints
Agentdebuggingrecord-repairer@1.0.0
Aautogenrecord-repairer
Record Repairer
Work the violation list field by field, mend only what the gate objected to, and send the record back for re-normalization instead of dropping it or writing it half-formed.
used in 0 blueprints
Tooldeploymentrecord-store@1.0.0
Aautogenrecord-store
Record Store
Upsert the accepted records into the target store in batches keyed on the upsert key, and return the commit receipt, row counts, keys written, and the batch the run can resume from.
#sql
used in 2 blueprints1 risk marker
Tooldeploymentrelease-gate@1.0.0
Aautogenrelease-gate
Release Gate
Admit only a signed-off build, write it to the run's release target with its tag and digest, and close the run.
used in 1 blueprint
Validationtestingremote-result-check@1.0.0
Aautogenremote-result-check
Remote Result Check
Judge a result this graph did not watch being produced, on nothing but the result itself and the task it answers.
used in 1 blueprint
Tooldeploymentreply-delivery@1.0.0
Aautogenreply-delivery
Reply Delivery
Send the reply that was cleared, or the one a reviewer released, and record that it went.
#messaging
used in 1 blueprint1 risk marker
Tooldeploymentreply-dispatch@1.0.0
Aautogenreply-dispatch
Reply Dispatch
Post the approved reply to the channel the ticket arrived on and return the provider's receipt, leaving the thread in whatever state that channel reports.
#messaging#http-fetch
used in 1 blueprint2 risk markers
Validationtestingreply-qa-check@1.0.0
Aautogenreply-qa-check
Reply QA Check
Grade the drafted reply against the tone, accuracy and completeness rubric, release it only above the confidence threshold, and emit the score the escalation path branches on.
used in 1 blueprint1 risk marker
Tooldeploymentreport-delivery@1.0.0
Aautogenreport-delivery
Report Delivery
Publish the grounded report to its reader with the citation trail intact, so every claim in it can be walked back to the passage that supports it.
used in 1 blueprint1 risk marker
Manager loopplanningresearch-supervisor@1.0.0
Aautogenresearch-supervisor
Research Supervisor
Turn the question into an agenda for one tournament and hold the round budget the rest of the graph spends.
used in 1 blueprint
Agentimplementationresolution-composer@1.0.0
Aautogenresolution-composer
Resolution Composer
Draft a candidate resolution for the routed request, grounding every claim in the reference document when one was retrieved and marking the gap when it was not.
used in 2 blueprints
Tooldeploymentresult-delivery@1.0.0
Aautogenresult-delivery
Result Delivery
Hand the accepted result to its destination and close the run, carrying the decision trace with it so the delivery can be reconstructed later.
used in 2 blueprints
Agentplanningretrieval-planner@1.0.0
Aautogenretrieval-planner
Retrieval Planner
Split the question into at most max_strands retrieval strands, one per corpus, and emit each as a query plan the strand can run on its own without asking the planner anything further.
used in 1 blueprint
Agentimplementationreview-drafter@1.0.0
Aautogenreview-drafter
Review Drafter
Turn the triage record into a review a maintainer can act on, one comment per finding, anchored to a file and a line, and fold the last test failure into the draft when the suite came back red.
used in 1 blueprint
Agentplanningreview-rubric@1.0.0
Aautogenreview-rubric
Review Rubric
State the standing bar a hypothesis has to clear, once, for the whole tournament.
used in 1 blueprint
Human gatereviewer-gate@1.0.0
Aautogenreviewer-gate
Reviewer Gate
Park anything a guard refused in front of a person, and let nothing past that the person has not released.
#human-review#messaging
used in 1 blueprint human in the loop
Validationtestingroute-critic@1.0.0
Aautogenroute-critic
Route Critic
Grade the answer, and separately record whether the lane it came from was the right one.
used in 1 blueprint
Agentplanningrubric-author@1.0.0
Aautogenrubric-author
Rubric Author
Turn the brief the run was instantiated with into an ordered, checkable list of criteria, and emit nothing else.
used in 1 blueprint
Agenttestingrun-summariser@1.0.0
Aautogenrun-summariser
Run Summariser
Aggregate the run's trace records into one report of per-step metrics: how many times each step ran, how many of those runs failed, and how long each took.
used in 1 blueprint
Tooldeploymentrunbook-executor@1.1.0
Aautogenrunbook-executor
Runbook Executor
Run the cleared mitigation's steps against the resolved targets, dry run first, then for real, and return the command log with every exit code in order.
#shell
used in 1 blueprint2 risk markers
Tooldeploymentrunbook-executor@1.0.0
Aautogenrunbook-executor
Runbook Executor
Run the cleared mitigation's steps against the resolved targets, dry run first, then for real, and return the command log with every exit code in order.
#shell
used in 0 blueprints2 risk markers
Toolrunbook-resolver@1.0.0
Aautogenrunbook-resolver
Runbook Resolver
Match the routed alert to a runbook in the versioned store and return it as Markdown, pinned to the revision that was current when the incident opened.
#vector-store#git
used in 1 blueprint
Shell toolsandboxed-python-runner@1.0.0
Aautogensandboxed-python-runner
Sandboxed Python Runner
Run the supplied script inside the sandbox and return what it printed, whatever that turns out to be.
#python-sandbox
used in 0 blueprints1 risk marker
Validationtestingschema-gate@1.1.0
Aautogenschema-gate
Schema Gate
Check the candidate record against the registered schema plus any assertions attached for this run, pass only what clears both, and emit the violation list for the rest.
used in 1 blueprint
Validationtestingschema-gate@1.0.0
Aautogenschema-gate
Schema Gate
Check the candidate record against the registered schema, pass only what conforms, and emit the violation list for everything that does not.
used in 1 blueprint
Agentplanningspec-planner@1.0.0
Aautogenspec-planner
Spec Planner
Turn the incoming request into two separate artefacts, an ordered build brief, and the acceptance criteria the finished work will be judged against.
used in 1 blueprint
Agentplanningstage-planner@1.0.0
Aautogenstage-planner
Stage Planner
Break the job into an ordered set of stages small enough to be re-run on their own, and mark the checkpoint boundary after each one.
used in 1 blueprint
Toolstep-tracer@1.0.0
Aautogenstep-tracer
Step Tracer
Turn every value a watched step emits into one trace record carrying the step's name, the time it arrived, its outcome, and the size and digest of the value, and hand the record on.
used in 1 blueprint
Agentdebuggingtargeted-debugger@1.1.0
Aautogentargeted-debugger
Targeted Debugger
Turn one run's failure evidence into the narrowest patch that addresses it, and stop once the attempt cap is spent or the evidence stops changing.
used in 1 blueprint
Agentdebuggingtargeted-debugger@1.0.0
Aautogentargeted-debugger
Targeted Debugger
Turn one run's failure evidence into the narrowest patch that addresses it, and stop once the attempt cap is spent or the evidence stops changing.
used in 0 blueprints
Agentplanningtask-decomposer@1.0.0
Aautogentask-decomposer
Task Decomposer
Break the request into an ordered set of sub-tasks small enough for one solver pass each, and carry the acceptance criteria down onto every one of them.
used in 1 blueprint
Tooltask-intake@1.0.0
Aautogentask-intake
Task Intake
Accept an inbound task request, stamp it with a run id and the acceptance criteria it will be judged against, and hand it to the line as a single self-contained brief.
used in 2 blueprints
Validationtestingtest-runner@1.1.0
Aautogentest-runner
Test Runner
Run the suites the review calls for against the PR head and gate on the result: green releases the report to the maintainer, red hands the failing cases back to the drafter for another pass.
#ci#git
used in 1 blueprint
Validationtestingtest-runner@1.0.0
Aautogentest-runner
Test Runner
Run the suites the review calls for against the PR head and gate on the result: green releases the report to the maintainer, red hands the failing cases back to the drafter for another pass.
#ci#git
used in 0 blueprints
Validationtestingtrajectory-auditor@1.0.0
Aautogentrajectory-auditor
Trajectory Auditor
Judge the route a run took rather than the answer it arrived at, and report where the route was wrong even when the answer was right.
used in 0 blueprints
Agentimplementationtransform-stage@1.0.0
Aautogentransform-stage
Transform Stage
Apply the plan's transformation to the staged working set, or to the snapshot the resume controller hands back, and emit the transformed set with the diff against what it started from.
used in 1 blueprint
Tooluntrusted-intake@1.0.0
Aautogenuntrusted-intake
Untrusted Intake
Admit inbound text from outside the system and hand it on typed as what it is: unscreened.
used in 1 blueprint1 risk marker
Toolvector-recall@1.0.0
Aautogenvector-recall
Vector Recall
Embed the strand's query and return the k nearest passages from the internal store, each carrying its document id and similarity score so the synthesizer can weigh how close it is.
#vector-store
used in 1 blueprint
Tooldeploymentwarehouse-publisher@1.0.0
Aautogenwarehouse-publisher
Warehouse Publisher
Advance the watermark to the end of the swept window, refresh the downstream views, and post the night's counts, rows cleaned, rows repaired, rows quarantined, to the ops channel.
#messaging
used in 1 blueprint1 risk marker
Toolweb-retriever@1.0.0
Aautogenweb-retriever
Web Retriever
Run the strand's query against public search, fetch the pages worth reading, and return each one as a passage stamped with its source URL and the time it was retrieved.
#web-search#http-fetch
used in 1 blueprint
Decisionimplementationweighted-vote@1.0.0
Aautogenweighted-vote
Weighted Vote
Score every competing proposal under the weighting rule and emit the winner together with the full tally, so the choice is auditable rather than a black box.
used in 1 blueprint
Agentplanningwork-planner@1.0.0
Aautogenwork-planner
Work Planner
Turn the objective into an ordered set of steps, without knowing how the result will be measured.
used in 1 blueprint