agentleFS
Sign inStart free
SkinnnyJaySkinnnyJay/wiki-llm/skills/wiki-retro/SKILL.md

wiki-retro

Skill0 starsChanged 4 months agoMITView on GitHub

Vault retrospective — activity, source mix, topic coverage, health, open questions, and recommendations from logs, git, and wiki-lint.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 1 is distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    If git.enabled, use llm-wiki git log / llm-wiki git lifecycle — not the parent repo’s .git for vault operations.

    line 14
  2. Very specific:

    Read wiki/log.md: count dated sections in the last 7 days (or window in $ARGUMENTS).

    line 20
  3. Very specific:

    If memory.enabled and raw/memory/ exists, count session files and note recent updated timestamps in frontmatter (wiki-session-memory).

    line 22
  4. Very specific:

    If raw/.preparation-log.jsonl exists, count prepare / raw finish actions in the window.

    line 27
  5. Very specific:

    Read wiki/index.md: list sections or topics with most new vs stale pages (use wiki/log.md dates when available).

    line 31
  6. Very specific:

    Run wiki-lint logic (or invoke the skill): orphans, broken [[wikilinks]], contradictions.

    line 35
  7. Distinctive:

    If vault git was enabled, commit subjects by phase are referenced when useful.

    line 62

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Wiki retro — Retrospective

Summarize research velocity, source mix, topic coverage, wiki health, and open questions from wiki/log.md, raw/.preparation-log.jsonl (if present), vault git history (if enabled), and a wiki-lint-style pass. Output is read-only analysis plus optional updates to wiki-learn memory.

Pre-flight

  1. Confirm llm-wiki/config.json exists. Resolve vault root (default llm-wiki/).
  2. If git.enabled, use llm-wiki git log / llm-wiki git lifecycle — not the parent repo’s .git for vault operations.

Steps

Step 1 — Activity summary

  • Read wiki/log.md: count dated sections in the last 7 days (or window in $ARGUMENTS).
  • Note lines matching research, ingest, pipeline, deep-research, etc.
  • If memory.enabled and raw/memory/ exists, count session files and note recent updated timestamps in frontmatter (wiki-session-memory).

Step 2 — Source mix

  • Scan raw/**/*.md (sample or full): group by source_type, platform, or path prefix (raw/news/, raw/academic/, …).
  • If raw/.preparation-log.jsonl exists, count prepare / raw finish actions in the window.

Step 3 — Topic coverage

  • Read wiki/index.md: list sections or topics with most new vs stale pages (use wiki/log.md dates when available).

Step 4 — Health

  • Run wiki-lint logic (or invoke the skill): orphans, broken [[wikilinks]], contradictions.
  • Compare to last retro if outputs/retro-*.md exists: delta one-liners.

Step 5 — Open questions

  • Aggregate Open questions: bullets from wiki/log.md and lint gaps.

Step 6 — Recommendations

  • Propose next research topics, pages to refresh, or integrations to enable.
  • Optionally append bullets under ## Next Steps in llm-wiki/.agent-memory.md (wiki-learn).

Step 7 — Write report

Save to outputs/retro-YYYY-MM-DD.md (create outputs/ if missing):

  1. Activity summary — sessions, ingests, wiki updates in window
  2. Source mix — breakdown by type/path
  3. Topic coverage — hot/cold areas
  4. Health trend — lint issues; delta vs last retro if known
  5. Open questions — collected list
  6. Recommendations — concrete next actions

Done looks like

  • outputs/retro-YYYY-MM-DD.md exists with all six sections (use N/A where data is missing).
  • User gets a short executive summary in chat plus path to the full report.
  • If vault git was enabled, commit subjects by phase are referenced when useful.

Artifacts

ReadsWrites
wiki/log.md, wiki/index.md, raw/**, raw/memory/**, raw/.preparation-log.jsonl, optional llm-wiki git logoutputs/retro-YYYY-MM-DD.md, optional llm-wiki/.agent-memory.md Next Steps

Related skills

  • wiki-lint — health signals
  • wiki-learn — persist recommendations
  • wiki-status — tooling/config snapshot
  • wiki-pipeline — standard flow for fixes
  • wiki-session-memory — per-session raw/memory/ files

Smoke check

  • CLI: From the vault root: llm-wiki validate; if vault git is enabled, llm-wiki git log -n 5.
  • Prompt: Invoke this skill; confirm retro uses wiki/log.md / git history when available.

Optional: persona.name in config (default Gennie).

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.

Check your setup, free