Skip to content

Latest commit

 

History

History

README.md

Exploratory missions

This directory holds mission skills that have been documented but not yet proven end-to-end. They live here, not in skills/, so the framework's shipped surface area maps 1:1 to missions with real-run evidence.

Why these are here

A mission stays in skills/ only if ALL three are true:

  1. A docs/<mission>-progress.md written from a real run on a real repo
  2. A docs/<mission>-readiness.md with a valid fleet-outcome block from the same run
  3. At least one external-repo run-archive — either under .fleet/runs/ in this repo with the mission named, or referenced from docs/external-dogfood/

The missions below fail at least one of those three checks, so they have been moved out of skills/. This is the same rule already applied to docs/exploratory/missions/scaffold-align/, docs/exploratory/missions/contract-first-build/, and docs/exploratory/missions/agents-layer/ in 2026-06-22, generalised to the post-Commit-C demotion of 2026-06-23.

The original demotion plan lives in docs/plans/way-ahead-2026-06-23.md § 3 Commit D.

Promotion criteria

A mission may move back to skills/ only when all three of the following artifacts exist and are referenced in the promotion PR:

Artifact Path
Progress doc docs/<mission>-progress.md (real run, not stub)
Readiness doc docs/<mission>-readiness.md with fleet-outcome block
External archive .fleet/runs/<run_id>/ produced by the mission, OR a referenced archive under docs/external-dogfood/

Doctrine alone is not sufficient. Tests inherited from autonomous-fleet-core are not sufficient. The promotion PR must cite a real coding-agent run that produced the archive.

Mechanical gate (CI): ./scripts/validate_mission_promotion.py --require-ready <slug> must exit 0 before moving a mission back to skills/.

Demoted missions (2026-06-23, Commit D)

bug-batch/

[Tier 2] Close a batch of bugs from a list/tracker/described set, one PR per bug, each gated by a FAILING TEST written first that reproduces the bug. If promoted, would force exactness on the agent-weakest category (bug-fixes). Currently: no progress doc, no readiness doc, no external archive.

cleanup/

[Tier 1] Behaviour-preserving code-health pass — dead-code removal, duplication kill, named anti-pattern fix — without re-architecting. If promoted, would be the lightest counterpart to the archived legacy-rebuild design. Currently: a readiness doc exists (docs/cleanup-readiness.md) but no progress doc and no external-repo run-archive. Demoted because the three-artifact rule requires BOTH progress AND external archive.

dependency-update/

[Tier 1] Update a repo's dependencies to current versions, fix bump breakages, keep the suite green, one PR per logical group. If promoted, would handle stale deps + security advisories autonomously. Currently: no progress, no readiness, no external archive.

design-integration/

[Tier 2] Adopt a fresh design across an existing product to full parity (visual AND feature-wise) — reskin every screen and build the features the design implies but the product lacks. If promoted, would subsume the narrower archived landing-page-convergence design. Currently: no progress, no readiness, no external archive.

inference-cost/

[Tier 2] Reduce AI inference spend while holding output quality constant — measurement-first cost optimization with a baseline cost+quality harness, sanctioned levers only, and a hard refusal of subscription-token-as-backend hacks. If promoted, would address one of the clearest measurable outcomes a mission can ship. Currently: no progress, no readiness, no external archive.

take-product-to-completion/

[Tier 3] Drive a STALLED product to a full-fledged, shippable state via adversarial review + market research + a FROZEN COHERENT PRODUCT BOUNDARY

  • full-depth build inside that boundary. If promoted, would be the flagship Tier 3 mission for finishing started-but-unshipped products. Currently: no progress, no readiness, no external archive.

targeted-migration/

[Tier 2] Migrate ONE axis of a codebase (framework version, library swap, language/runtime bump, DB/ORM change, API-version move) while preserving everything else and keeping the suite green. If promoted, would handle the one-axis-at-a-time slot between dependency-update and the archived legacy-rebuild design. Currently: no progress, no readiness, no external archive.

gstack-derived missions (2026-06-27)

Mapped from garrytan/gstack specialist skills into fleet frozen-ledger missions. Research: docs/gstack-missions-research.md.

browser-qa-fix/

[Tier 2] Browser-grounded QA with screenshot EVID and fix loop until health threshold met. gstack sources: qa, qa-only, browse, health.

incident-investigate/

[Tier 2] Root-cause analysis with frozen docs/incident-rca.md and mandatory regression test. gstack sources: investigate, retro, learn.

Archived

Some former exploratory missions are parked under archive/. They are excluded from the active exploratory set until they satisfy the standard evidence triple and, where they overlap another mission, a written differentiation argument. See archive/README.md.

Promotion process

To re-promote a mission from docs/exploratory/missions/<mission>/ back to skills/<mission>/:

  1. Run the mission on a real repo via a campaign (in this repo or an external one). The mission must produce a valid fleet-outcome block.
  2. Archive the run to .fleet/runs/<run_id>/ with a passing validate_run_archive.py. If the run was on an external repo, reference it under docs/external-dogfood/<mission>-<repo>.md.
  3. Write docs/<mission>-progress.md documenting the run end-to-end (the actual builder/reviewer transcript, the findings closed, the metrics produced).
  4. Write docs/<mission>-readiness.md with the fleet-outcome block from the run.
  5. git mv docs/exploratory/missions/<mission> skills/<mission> to restore the skill.
  6. Remove status: exploratory from the SKILL.md frontmatter and strip the exploratory admonition block at the top of the file.
  7. Update consumers — re-add the mission to skills/autonomous-fleet/SKILL.md, skills/autonomous-fleet/references/missions.md, README.md's mission list, and any scripts/campaigns/*.yaml that should re-include it.
  8. Update the marketplace packet to mention the newly promoted mission.
  9. Open a PR citing the run id, the archive path, the progress doc, and the readiness doc.

A demotion can be reversed; doctrine alone cannot promote. The artifact is the gate.