/distill

name: distill
description: "One-time snapshot extracting patterns from work history and accumulated lessons, distills into concrete improvements — new agent/skill suggestions, memory pruning, consolidating lessons into rules/agent updates, or performing bin/ extraction from /audit --efficiency candidates. Roster boundary analysis → /foundry:audit agents (Check 34)."
argument-hint: '[prune | memory | executables [<run-dir-or-report-path>] | "external <url-or-path>" | "<recurring task description>"] [--project] [--eager] [--keep "<items>"]'
disable-model-invocation: true
allowed-tools: Read, Edit, Bash, Glob, Grep, Write, AskUserQuestion, Agent, WebFetch, TaskCreate, TaskUpdate, TaskList
effort: low

<objective>

Analyze how Claude Code is used and surface concrete improvements — new agents/skills to reduce repetition, or consolidate lessons into governance files (rules, agent instructions, skill updates) — without duplicating what exists.

NOT for single-file edits or quality checks — use `/foundry:audit` for config quality checks. NOT for audit-only scan for extraction candidates (use `/foundry:audit --efficiency` instead of `distill executables` for detection-only).

</objective>

<inputs>

• **$ARGUMENTS**: optional. Modes:
• Omitted — analyze existing patterns and agents; generate suggestions proactively.
• `prune [--eager]` — evaluate project memory file for stale, redundant, or verbose entries. Default: advisory diff + apply prompt. `--eager`: score every entry (Usage likelihood × Impact → Tier P0/P1/P2), print full scored table with `#` column, let user select by tier or item numbers, delegate edits to `foundry:curator`.
• `memory [--eager]` — read `.notes/lessons.md` and memory feedback files, distill recurring patterns into proposed rule files, agent instruction updates, and skill workflow changes. `--eager`: include Pattern count, Strength, and Tier columns in proposal table; let user select clusters to promote by tier or item numbers; delegate writes to `foundry:curator`.
• `external <source> [--eager]` — analyse external plugin, skill, or agentic resource and produce structured adoption proposal. `<source>` is URL, file path, or local directory. `--eager`: lower adoption bar — recommend partial adoption even for single useful components.
• `executables [--eager] [<run-dir-or-report-path>]` — perform bin/ extraction from `/foundry:audit --efficiency` Check 33 candidates. Auto-detects latest run dir under `.reports/audit/`; pass optional path to target a specific run dir or report file. Runs inline Check 33 scan when no report exists. Default gates on HIGH/MEDIUM verdict. `--eager`: also surface LOW verdict clusters as extraction candidates. Spawns `foundry:sw-engineer` per cluster. Skip to **Mode: Executables Extraction** below.
• `[--eager] <recurring task description>` — use description as context when generating suggestions. `--eager`: lower frequency threshold from 3+ to 2+ occurrences; single high-effort occurrence also qualifies.
• `--project` — in `prune` and `memory` modes, show an interactive project picker: enumerate all slugs under `~/.claude/projects/*/memory/` with MEMORY.md size in tokens, then let user select which project(s) to operate on. Omit to operate across **all** projects automatically. Has no effect on other modes.

</inputs>

<compaction> Key boundary 1: end of Step 2 frequency heuristics (default mode only — prune/memory/external/executables exit early), before Step 3 gap analysis. Preserve at boundary 1: EAGER flag, ARGUMENTS (stripped), no run-dir (default mode is stateless). Terminal paths: end of Step 5 report (default mode); end of Memory Pruning mode (both eager and standard branches). </compaction>

<workflow>

**Task hygiene**: load and follow the protocol below.

# loads: compaction-contract.md
# audit-skip: resilience-replication
_FS=$(python "${CLAUDE_PLUGIN_ROOT:-plugins/cc_foundry}/bin/resolve_shared_path.py" foundry skills/_shared 2>/dev/null || echo "plugins/cc_foundry/skills/_shared")  # timeout: 5000
cat "$_FS/task-hygiene.md"

export CSID="${CLAUDE_CODE_SESSION_ID:-$PPID}"
KEEP_ITEMS=""
if [[ "$ARGUMENTS" =~ --keep[[:space:]]\"([^\"]+)\" ]]; then
    KEEP_ITEMS="${BASH_REMATCH[1]}"
fi
ARGUMENTS=$(echo "$ARGUMENTS" | sed 's/--keep "[^"]*"//g')
rm -f .temp/state/skill-contract.md  # clear stale contract (compaction-contract.md §Lifecycle)  # timeout: 5000
mkdir -p "${TMPDIR:-/tmp}/distill-state-${CSID}"
echo "$KEEP_ITEMS" > "${TMPDIR:-/tmp}/distill-state-${CSID}/keep-items"
EAGER=false
[[ "$ARGUMENTS" == *"--eager"* ]] && EAGER=true
ARGUMENTS=$(echo "$ARGUMENTS" | sed 's/--eager//g' | xargs)  # timeout: 3000
echo "EAGER=$EAGER"  # shell vars don't persist across Bash calls — read from stdout
echo "ARGUMENTS_STRIPPED=$ARGUMENTS"

> **Note**: `EAGER` and stripped `ARGUMENTS` are set by this Bash block, but shell variable state does **not** persist across separate Bash() tool calls. After this block runs, read its stdout (`EAGER=true/false`, `ARGUMENTS_STRIPPED=...`) and carry those values as model-context references for all subsequent mode dispatch and threshold decisions. Do not rely on `$EAGER` as a live shell variable in later steps — substitute the literal boolean value read from stdout.

PROJECT_FLAG=false
if echo "$ARGUMENTS" | grep -qE -- "--project"; then
    PROJECT_FLAG=true
    ARGUMENTS=$(echo "$ARGUMENTS" | sed 's/--project//' | xargs)
fi
echo "PROJECT_FLAG=$PROJECT_FLAG"
echo "ARGUMENTS_FINAL=$ARGUMENTS"

> **Note**: `PROJECT_FLAG` does not persist across Bash calls. Read its value from the stdout line `PROJECT_FLAG=true/false` and carry as model-context reference. When `true`, the mode must run the interactive picker before operating.

Step 1: Inventory existing agents and skills

Use Glob tool to enumerate agents and skills across all sources — project-local AND plugin-namespaced — to avoid false-gap findings when candidate already exists in plugin:

• **Project-l