bounded-autonomy
Skill2 starsChanged 32 days agoMITView on GitHub
Applies Captain-approved routing weight proposals under autonomy budget (Level 3 bounded)
In this repository52 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .cursor/rules/00-core-operating-model.mdcCursor rule
- .cursor/rules/01-plan-approval-gate.mdcCursor rule
- .cursor/rules/02-git-worktree-policy.mdcCursor rule
- .cursor/rules/03-validation-definition-of-done.mdcCursor rule
- .cursor/rules/04-documentation-memory.mdcCursor rule
- accessibility-reviewSkill
- agent-skills-packagingSkill
- autonomy-budgetSkill
- candidate-promotionSkill
- capability-planningSkill
- code-reviewerSkill
- code-structure-cleanupSkill
- compass-evaluatorSkill
- context-inodesSkill
- dependency-supply-chainSkill
- docker-cloudSkill
- embedding-providersSkill
- execution-telemetrySkill
- experience-routingSkill
- experience-skill-trainingSkill
- external-knowledge-ingestSkill
- github-integrationSkill
- goose-agent-orchestrationSkill
- harness-gcSkill
- hosted-vector-dbSkill
- implementation-planningSkill
- ios-engineeringSkill
- knowledge-stewardSkill
- linear-integrationSkill
- node-engineeringSkill
- northstar-connected-routineSkill
- notion-integrationSkill
- package-registry-tiSkill
- persistent-role-promotionSkill
- playwright-browser-validationSkill
- postgres-prismaSkill
- procedure-playbooksSkill
- project-overseerSkill
- pull-request-preparationSkill
- python-mlSkill
- react-engineeringSkill
- repository-discoverySkill
- review-fix-loopSkill
- security-audit-proceduresSkill
- security-reviewSkill
- skill-learning-loopSkill
- skill-lifecycleSkill
- source-code-contextSkill
- technology-intelligence-liveSkill
- testing-validationSkill
- worktree-orchestrationSkill
The file
Files
Bounded Autonomy
Use this Skill when
The Captain has reviewed an experience-routing proposal and wants to apply matcher weight suggestions under an autonomy budget (Milestone 4 Level 3).
Inputs
- Routing proposal JSON under
.agent/routing/proposals/withcaptain_approved: true(Captain sets this flag explicitly) - Autonomy budget ledger path (recommended)
Procedure
- Generate or locate a proposal (
experience-routingSkill). Confirmauto_applyremainsfalse. - Captain edits the proposal to set
"captain_approved": true. Apply once under budget:
./scripts/apply-routing-proposal.sh \ --proposal .agent/routing/proposals/<id>.json \ --budget .agent/budgets/<plan-id>.mdNorthStar connected routine may surface pending proposals after
REVIEW_READY(--surface-routing) but never applies weights from plan approval alone. See Skillnorthstar-connected-routine.Context selection apply (M17 — which plan slices appear):
./scripts/propose-context-selection.sh --experiences tests/fixtures/experience/contact-counter.json # Captain sets captain_approved: true on the context-selection proposal ./scripts/apply-context-selection-proposal.sh \ --proposal .agent/routing/proposals/<ctxsel-id>.json \ --budget .agent/budgets/<plan-id>.md- Review audit under
.agent/routing/applied/and re-run evals. - Rollback if needed by restoring
orchestrator/matcher/weights.jsondefaults (orrollback/pre-m4-persistent-roles-bounded-autonomy).
Output
- Updated
orchestrator/matcher/weights.json - Audit record under
.agent/routing/applied/ - Incremented weight-apply counter on the budget ledger
Prohibited actions
- Applying without
captain_approved: true - Auto-applying from proposal generation
- Mutating Skills or
.cursor/agents/from this Skill - Exceeding the plan's maximum weight-apply operations
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

