agentleFS
tigusigalpatigusigalpa/ai-kit/.agents/skills/project-continuity/SKILL.md

project-continuity

Skill25 starsChanged 3 days agoMITView on GitHub

Maintain AI-KIT's own context, template boundaries, and installer verification when kit structure or procedures change.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. locez/locez-skills★ 19same name
    2. fugui6688661/glom-continuity★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository9 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Keep the root context/history about this kit.

    line 9
  2. Distinctive:

    Check candidate draft preservation, state/settings validation, and adapter retirement conflicts when changing installation.

    line 15
  3. Distinctive:

    Do not bootstrap this kit as an application or infer runtime success from static checks.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

AI-KIT maintenance continuity

Read contribution rules. Read optional local ../../../PROJECT_CONTEXT.md, ../../../AGENTS.md and ../../../WIKI.md when present; otherwise verify facts from public source/configuration. These three root files stay local and untracked, outside wheel/Docker payloads; their template counterparts stay public. Check distributions with and without local copies and preserve their on-disk bytes when untracking. Keep the root context/history about this kit; keep template project facts uninitialized. Sync commands, this skill, docs, context and WIKI on procedure/fact/agreement changes; keep facts in context. Keep release receipts (commits, workflow runs, hashes, digests) in docs/V*.md and link them; context holds current facts, commands and agreements only. Validate templates/links/registry, the agent registry (integrations/agents.json), and behavioral installation scenarios; record root CHANGELOG entries. Keep the root SKILL.md (the repository as a skill) consistent with the CLI, presets, extras, and policies. Review explicit semantic acceptance and future-upgrade conflicts whenever installer baseline behavior changes. For model/instruction controls, exercise all four combinations, absent-key legacy defaults, flag/preset independence, saved upgrades, native model/effort preservation and custom-helper conflicts. Minimal must retain mandatory/nested architecture, security and checks above context budgets. Disabled model advice must gate route/adaptive/proposals and managed helper overrides without claiming a host lock. Keep configured/requested/observed values distinct; parent runtime telemetry unavailable to the offline CLI stays null. See audit and contract. For model exceptions, cover family/version syntax versus exact IDs, global-off precedence, unknown/current declarations, no capability inference, persistent append/clear/idempotency, and invocation-only preservation with zero writes. Recheck all managed helper files, retained Aider/native overrides, unchanged user effort, and benchmark fixed-route/exception identity. Keep optional lists empty by default; inherited host defaults and already loaded native settings remain external limitations. Check candidate draft preservation, state/settings validation, and adapter retirement conflicts when changing installation. Exercise installed Git visibility and portable linked-path refusal; distinguish synthetic fixtures from real application trials. Verify the conditional Docker trigger in installed AGENTS and preserve edited entry points during upgrades. Exercise file-versus-directory native rule layouts on both probe and selected-write paths: simulate POSIX ENOTDIR on Windows, preserve existing file markers and rerun stability, and refuse blocked selected parents before writes. Do not hide permission failures or linked-path refusals during optional adapter probes; see file-parent compatibility. Exercise manifest-based profile detection and project-fact drafts (version/commands into PROJECT_CONTEXT and project.json), scoped Node/Python ignore rules, the installed-project doctor report, opt-in extras (session-start, guards, CI) including key-level client settings merges, legacy whole-file migration, and hook execution, client detection markers, interactive/check onboarding, the adr/changelog helpers, and the offline router helper (route/configure with provider selection) when installer behavior changes. For generated client configuration, exercise private/team settings files and mode switches, owned-entry moves, project.json precedence, secret lists, ignore blocks, every scoped rule format with root modules and skipped paths, presets that turn extras off, and workflow skill copies; recheck client formats against primary docs before changing them. For routing, compare the implementation ladder with canonical POLICY, exercise English/Russian operation/risk regressions and override precedence, provider-review status/sources, and distinguish recommendations from executed model controls. For stack profiles, exercise manifest evidence, module scoping, and command drafting; for Filament specifically test Composer dependency/lock evidence and source/native skill lifecycle. For optional metrics, verify failure/recovery accounting, missing-value coverage, duplicate/lock/path guards, private/team journal visibility, schema 1/2 compatibility, experiment/arm/task identity, per-provider analyze output, and upgrade preservation. Exercise context snapshots and doctor warnings for new/removed modules, profile drift, and manifest changes. Exercise adapter reports plus explicit retirement of unchanged files, backups/state updates, and modified-file conflicts. Exercise opaque evaluator pair coverage without task-text persistence. Use the evaluation protocol for separately authorized application trials; synthetic fixtures establish no savings. For language/identifier/client changes, verify fresh English and preserved saved languages, JavaScript/Node.js versus browser evidence, cross-stack --with-uuid opt-in, and --with-claude-github persistence without login/Git mutation. Claude commit attribution is an owned singleton scalar; preserve custom-value conflicts, human identity/signing and unrelated PR settings across private/team moves. Test Devin (Windsurf) .devin rules, legacy .windsurf coexistence and explicit retirement without deleting adapted files. Keep TypeUI setup in the README source-checked against vendor MCP/auth/scope docs; installation does not establish live activation. Record runtime gaps separately from static/source checks. For measurement-led changes, test schema 1/2/3 compatibility, fixed A/B controls, immutable Git blobs and isolated repeated workspaces, missing/estimated telemetry, semantic-review gates, interrupted runs and idempotent private exports. Exercise resume identity checks, per-unit failure recording and configurable kit-arm agents/presets. Check that worker Git discovery cannot reach the source through parents: a ceiling at cwd itself is ineffective. Keep worker requests separate from observed execution. Exercise routing modes/pins/profiles, distinct-case quality floors, matched history, unsupported controls, immutable proposal history and explicit backed-up restore. Exercise module context scope, duplicate content, critical rules/ADRs above byte budgets and linked paths. Follow the staged plan and worker contract; keep native activation and real trials separate. For optional orchestration, test opt-in/persistence, Codex helper conflicts/retirement, private/team/archive exclusions, bounded inputs/keys/budgets, reservations/locks, unknown billing, source-bound review and reconciliation without replay. Keep native costs separate; tests make no live calls. Resolve temporary fixture roots before deriving paths used by guards, assertions, or injected failures. Test equivalent path spellings; keep cleanup bound to original disposable directories. Windows CI regression records the relevant scenario. For reference upgrades verify the published tree and overlay leftovers; preserve canonical links and checker coverage. For aikit mcp report, recheck client MCP file formats against vendor docs before changing them, assert that no fixture secret appears anywhere in the serialized report, and cover Python 3.10 (no tomllib). Exercise the downgrade guard in preview, apply and doctor --fix. After changing installer options, regenerate integrations/installer-flags.json with aikit flags --output integrations/installer-flags.json; kit validation compares it, the flag guide and SKILL.md with argparse. Review selection precedence by hand; check Windows examples and actual client activation. For OpenRouter documentation, recheck primary API/cache/usage docs; distinguish shipped context manifests from proposed request assembly/runtime, prompt caching from response reuse, and observed usage from assumed savings. README/Overview are unbounded; keep instruction budgets and link/language checks. For PyPI releases, verify metadata, VERSION, licenses, description, resources and private/path exclusions; build the wheel from sdist and smoke-install offline. Quote Twine globs; report OS gaps. A JSON 404 does not prove name eligibility. Test release/manual identity guards without uploads; preserve CI names and old tags. Published stable releases prepare PyPI artifacts; manual runs default to validation. Require exact version/tag/repository gates, owner approval and future tag coverage; agents must not approve deployments. OIDC belongs only in upload jobs consuming validated same-run artifacts. Keep setup, upload and live MCP activation separate. For CI Python bootstrap retries, preserve the upstream SHA, bounded waits, cancellation guards and final hard failure. Never retry tests/build/upload as setup recovery. Verify same-commit local-action loading, read-only checkout, unchanged required check names and wheel/Docker inclusion; distinguish static guards from native hosted retry execution. For repository administration or changes to existing required CI names, consult repository maintenance, verify live protection and emitted names, and keep required checks aligned. New local jobs establish no hosted result or required-check update. Remote changes need explicit authorization; Core governs commits/pushes. Assess Docker impact on every task, including installed guidance. Rebuild/test affected final sources as ai-kit:dev or record a descriptive-only exception. Verify client/daemon, pins, VERSION/labels, closed context, wheel resources, non-root mounts, offline/read-only preview/apply, preservation, reruns, signals and Job schema. Follow canonical cleanup: no threshold, zero eligible disposable resources, dedicated-builder cache, protected persistent/unrelated data and separately measured RAM/disk reclamation. Check actual backend, VHDX identity and idle maintenance scope; never infer compaction from prune. Upgrade tests must preserve adapted container rules/AGENTS through candidates. Record actual runtime gaps; publication needs its authorization. For optional task workflows, verify default absence/independent opt-in, created-once examples and private-state preservation across presets, mode switches and upgrades. Exercise create/revise idempotency, explicit transitions/cancellation/reopening, read-only previews/status, atomic locks/pending runs, Git HEAD/index/dirty/untracked/deleted source fingerprints, scope/link/submodule/credential/budget refusals, failed/timeout/interrupted/source-mutating checks, exact-fingerprint criterion acceptance and stale accepted reports. Keep .tasks state out of Git, archives, wheel, image and link/context scans. Run the actual packaged CLI in wheel/Docker checks. Local assertions are not authenticated human review, commands are not sandboxed, and ignored dependencies/tool versions are outside source fingerprints. Automatic coordination and real model execution remain separate. For dialogue memory, test all supported imports, selected ChatGPT branches, omissions, untrusted roles, pending prose, explicit keyed conflicts/resolutions, source-bound review, immutable reruns, both exports and byte limits. Exercise linked/blocked paths, private/team upgrade preservation, Git/build/package exclusions and installed CLI execution. Do not infer semantic agreement from no keyed conflicts or claim live sync/upload from local export. Preserve raw sources and default project context. For optional upstream skills, follow security-audit and Strix integration: preserve pinned bytes/license and newline attributes, verify bounded tree coverage/digests and nested-path aliases, explicit-only metadata, absent default/preset payloads, independent persistent choices, local adaptations/conflicts, links, visibility and wheel resources. Exempt only verified external examples from authored model/language/budget checks. Keep installation separate from audits/runtime setup; report unsupported validators. For Strix, check writable copy isolation, explicit full/diff scope, budget/data destination and no implicit cloud fallback; integration tests must start no scan, model or container. For Context7, check opt-in skill/registry selection, independent persistence, source/native copies, local conflicts and unchanged native MCP configuration. Keep normal skill discovery, version-aware sanitized queries and remote activation separate. Recheck client schemas/authentication against primary docs; a source revision does not pin the hosted service. For Web Design, check whole-module default absence, multi-entry skill selection, independent persistence, created-once project documents, native copies/retirement, local conflicts and linked/blocked paths. Exercise workflow context scope, required design memory above budgets, plain HTML without assumed stack facts, private raw output/team visibility and offline wheel resources. Keep TypeUI/SkillUI setup and real browser/Laravel acceptance separate from installer evidence; adopt external guidance only within the selected stack/visual system. For the English/Devin/UUID/GitHub identity release use v1.6.0 for receipts; keep policy in its canonical templates. For the optional task workflow release use v1.7.0 for receipts; keep behavioral policy in the task contract. For conditional exception releases use v1.9.0 for receipts; preserve published artifacts and unknown-model provenance. For model/instruction control releases use v1.8.0 for receipts; keep behavior in Engineering and settings, and distinguish platform telemetry from configured models. For a requested release, synchronize VERSION, root changelog, README, WIKI and maintainer context; keep kit history out of templates. Recheck CLI/version, kit, isolated archives, offline installation and prior-version upgrade preservation. Distinguish checks before/after labeling. Follow the release path: preserve tags, wait for hosted build checks and report pending human approval; never approve PyPI for the owner. Git authorization follows Core. Release notes must distinguish installed guidance from external tools, live client/model setup and generated-site acceptance; verify security examples against pinned contracts. Keep every distribution byte-identical: .gitattributes pins LF (vendor bytes stay -text), Docker smoke rejects CR in packaged text, and images are built from a clean git archive export with AIKIT_VERSION and AIKIT_REVISION. Upgrades must treat line-ending-only differences as unchanged. Keep the PyPI build job on hash-pinned tools, twine in its own job and digest verification before upload; pin conditional build-tool dependencies (such as tomli below Python 3.11) and run build-tool changes on Python 3.10 in a container before pushing, because local Windows runs use a newer Python; tests/test_workflow_security.py guards these properties and pinned action SHAs. Tracked root entries (CLAUDE/KIMI/MANUS) must work without local maintainer files. The project CI template validates committed files only. Keep docker_cleanup modes (maximum default, keep-cache) consistent across CONTAINERS, AGENTS, settings, installer and docs. For every Docker image release, follow the Docker Hub checklist and update the complete Overview; synchronize version/features/examples before building and verified publication identity afterward. For Docker releases, follow image verification; keep local image identity distinct from a registry digest and source-bound human review distinct from automatic conversation synchronization. After explicitly authorized publication, read back the registry digest, platform and version/configuration; preserve the prepared image and record later publication-only documentation separately. Do not assign additional tags implicitly. Do not bootstrap this kit as an application or infer runtime success from static checks.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.