agent-wiki-compare-out…
Compare successful and failed normalized agent trajectories to derive evidence-backed agent-wiki guidelines. Use when Codex has…
Must be used at the start of any non-trivial task involving code changes, debugging, repo exploration, file inspection, or environment/tooling investigation to surface stored guidance before analysis or tool use.
Claude Code
$ npx -y skills add AgentToolkit/altk-evolve --skill recall --agent claude-code
Codex
$ npx skills add AgentToolkit/altk-evolve --skill recall -a codex -y
Cursor
$ npx skills add AgentToolkit/altk-evolve --skill recall -a cursor -y
Gemini CLI
$ npx skills add AgentToolkit/altk-evolve --skill recall -a gemini-cli -y
GitHub Copilot
$ npx skills add AgentToolkit/altk-evolve --skill recall -a github-copilot -y
OpenCode
$ npx skills add AgentToolkit/altk-evolve --skill recall -a opencode -y
Windsurf
$ npx skills add AgentToolkit/altk-evolve --skill recall -a windsurf -y
name: recall description: Must be used at the start of any non-trivial task involving code changes, debugging, repo exploration, file inspection, or environment/tooling investigation to surface stored guidance before analysis or tool use.
This skill loads relevant stored Evolve entities into the current turn before substantive work begins.
Use this skill first whenever the task involves:
Skip only for trivial conversational requests with no local context.
Before any non-trivial local work, you must complete the recall workflow below. Reading this `SKILL.md` alone does not satisfy the skill.
Do not proceed to other analysis or tool use until all steps below are complete.
1. If a manifest has already been injected for this turn, use it to pick which entity files to open. Otherwise inspect `${EVOLVE_DIR:-.evolve}/entities/` and `${EVOLVE_DIR:-.evolve}/public/` for guidance relevant to the current task. 2. Read each matching entity file that appears relevant. 3. Summarize the applicable guidance in your own words before proceeding. 4. If no relevant entities exist, state that explicitly before proceeding.
Before moving on, produce an explicit completion note in your reasoning or user update using one of these forms:
1. List or search files under `${EVOLVE_DIR:-.evolve}/entities/` and `${EVOLVE_DIR:-.evolve}/public/` (or read the injected manifest if one is present). 2. Identify candidate entities relevant to the task. 3. Open and read those entity files. 4. Summarize what applies, or state that nothing applies.
The skill is not complete if any of the following are true:
1. The Claw-code `PreToolUse` hook fires before each tool call. 2. The helper script reads tool input from stdin (best-effort, ignored beyond logging). 3. It emits a minimal manifest from `${EVOLVE_DIR:-.evolve}/entities/` and `${EVOLVE_DIR:-.evolve}/public/` containing only `path`, `type`, and `trigger`. 4. Claw-code uses that manifest to decide which full entity files to read on demand. 5. If the hook is not active, this skill remains the full manual fallback: inspect the entity files directly, read the relevant ones, and summarize what applies.
.evolve/entities/
guideline/
use-context-managers-for-file-operations.md <- private
subscribed/
memory/ <- write-scope clone (publishes land here)
guideline/
my-published-guideline.md
alice/ <- read-scope clone
guideline/
alice-guideline.md <- annotated [from: alice]Automatic hook output is manifest-first. Each manifest entry contains only:
{"path": ".evolve/entities/guideline/use-context-managers-for-file-operations.md", "type": "guideline", "trigger": "When processing files or managing resources"}Each file still uses markdown with YAML frontmatter:
--- type: guideline trigger: When processing files or managing resources --- Use context managers for file operations ## Rationale Ensures proper resource cleanup
Evolve helps AI agents learn from experience. It turns past trajectories into reusable guidelines that help agents complete future tasks, without changing model weights.
Compare successful and failed normalized agent trajectories to derive evidence-backed agent-wiki guidelines. Use when Codex has…
Read all atomic guidelines in wiki-twobatch/guidelines/ and propose themed clusters that group near-duplicates. Writes cluster…
Consult an agent-wiki for guidelines relevant to the task at hand. The wiki itself documents how to retrieve from it (AGENTS.md).…
Read a normalized Claude Code trajectory JSON and extract reusable guidelines into wiki-twobatch/guidelines/. Use when mining…
Ingest one or more agent trajectories (raw bob/claude traces or normalized JSON) into an agent-wiki end-to-end — convert,…
Read a normalized Claude Code trajectory JSON and write an episodic summary page to wiki-twobatch/summaries/. Use when…
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
Ask which skill or flow fits your situation. A router over the skills in this repo.