Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/ashlrai/ashlr-plugin/ashlr-genome-loopgit clone --depth 1 https://github.com/ashlrai/ashlr-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/ashlrai/ashlr-plugin/ashlr-genome-loop)<a href="https://agentmods.dev/commands/ashlrai/ashlr-plugin/ashlr-genome-loop"><img src="https://agentmods.dev/badge/commands/ashlrai/ashlr-plugin/ashlr-genome-loop.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00027 | $0.00954 |
| Opus 5 | $0.00014 | $0.00477 |
| Sonnet 5 | $0.00005 | $0.00191 |
| Haiku 4.5 | $0.00003 | $0.00095 |
Grade A, and why
ashlr-genome-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The ashlr-plugin runs an automatic genome loop: a PostToolUse hook proposes
architecture/decision-flavored observations to
.ashlrcode/genome/proposals.jsonl, and a SessionEnd hook consolidates them
back into section files when ≥ 3 are pending. This command exposes the
loop's status and on/off state.
Subcommands (arg to the slash command):
status— show pending proposal count, last consolidation timestamp, and whether the loop is currently enabled.run— force a consolidation pass on the current project's genome.on— enable the loop (writes~/.ashlr/config.json).off— disable the loop (writes~/.ashlr/config.json). TheASHLR_GENOME_AUTO=0env var also disables it per-shell.
Dispatch on $1:
If $1 is empty or status
Run the following Bash command and relay the output:
bash -c '
set -e
CFG="$HOME/.ashlr/config.json"
LOG="$HOME/.ashlr/genome-consolidation.log"
GENOME="$(pwd)/.ashlrcode/genome"
PROPOSALS="$GENOME/proposals.jsonl"
ENABLED="on"
if [ "${ASHLR_GENOME_AUTO:-1}" = "0" ]; then ENABLED="off (env ASHLR_GENOME_AUTO=0)"; fi
if [ -f "$CFG" ] && grep -q "\"genomeAuto\"[[:space:]]*:[[:space:]]*false" "$CFG" 2>/dev/null; then
ENABLED="off (~/.ashlr/config.json)"
fi
PENDING=0
if [ -f "$PROPOSALS" ]; then
PENDING=$(grep -c . "$PROPOSALS" 2>/dev/null || echo 0)
fi
LAST="(never)"
if [ -f "$LOG" ]; then
LAST=$(tail -n 1 "$LOG" 2>/dev/null | awk "{print \$1}")
[ -z "$LAST" ] && LAST="(never)"
fi
echo "ashlr genome loop — $ENABLED"
echo " pending proposals: $PENDING ($PROPOSALS)"
echo " last consolidation: $LAST"
'
If $1 is run
Run the following Bash command and relay its output verbatim:
bun run ${CLAUDE_PLUGIN_ROOT}/scripts/genome-auto-consolidate.ts --dir "$(pwd)"
If no output appears, say "No pending proposals to consolidate (need ≥ 3)."
If $1 is on
Run the following Bash command:
bash -c '
mkdir -p "$HOME/.ashlr"
CFG="$HOME/.ashlr/config.json"
if [ -f "$CFG" ]; then
bun -e "
const fs = require(\"fs\");
const p = process.env.HOME + \"/.ashlr/config.json\";
let j = {};
try { j = JSON.parse(fs.readFileSync(p, \"utf-8\")); } catch {}
j.genomeAuto = true;
fs.writeFileSync(p, JSON.stringify(j, null, 2));
console.log(\"ashlr genome loop: enabled\");
"
else
echo "{\"genomeAuto\": true}" > "$CFG"
echo "ashlr genome loop: enabled"
fi
'
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 119 lines · 27 tokens per session scan A 4b0dffef809d
ashlr-genome-loop is a command published in the GitHub repository ashlrai/ashlr-plugin (3 stars, last pushed 4d ago), licensed MIT. It adds 27 tokens to every session and 954 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
piratebao-compress
Compress a natural-language memory file.
nimbus-implicit-knowledge-extraction
The two implicit-knowledge extraction pipelines behind nimbus glossary and nimbus decisions (Spine S1 "Local Brain") — packages/gateway/src/glossary/ and packages/gateway/src/decisions/. Covers the shared four-stage pass shape (discover → score/prioritize → LLM consolidate/extract → reconcile), the : source-type…
cleanup-context
Memory optimization - removes duplicates, consolidates knowledge, and cleans memory banks for better performance.
lessons
View project lessons in MEMORY/LESSON.md (project root) and append a compact "never reproduce" lesson.
memory
Neural memory operations - search, store, stats, consolidate.
minutes-ideas
Surface recent voice memos and ideas captured from any device. Use when the user asks "what ideas did I have?", "what were my recent memos?", "what did I record while walking?", or wants to recall a captured thought.