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/zircote/git-notes-memory/recallgit clone --depth 1 https://github.com/zircote/git-notes-memoryWhat 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 | $0.00011 | $0.01328 |
| Opus 5 | $0.00005 | $0.00664 |
| Sonnet 5 | $0.00002 | $0.00266 |
| Haiku 4.5 | $0.00001 | $0.00133 |
Grade A, and why
recall 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 yesterday.
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 — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<help_check>
Help Check
If $ARGUMENTS contains --help or -h:
Output this help and HALT (do not proceed further):
<help_output>
RECALL(1) User Commands RECALL(1)
NAME
recall - Recall relevant memories for the current context or a s...
SYNOPSIS
/memory:recall [query] [--namespace=ns] [--limit=n]
DESCRIPTION
Recall relevant memories for the current context or a specific query
OPTIONS
--help, -h Show this help message
EXAMPLES
/memory:recall
/memory:recall <query>
/memory:recall --help
SEE ALSO
/memory:* for related commands
RECALL(1)
</help_output>
After outputting help, HALT immediately. Do not proceed with command execution. </help_check>
/memory:recall - Recall Relevant Memories
Retrieve relevant memories from the git-backed memory system.
Your Task
You will help the user recall memories relevant to their current context or query.
Arguments format: $ARGUMENTS
Parse the arguments:
- Extract
--namespace=<ns>if present (one of:decisions,learnings,blockers,progress,reviews,patterns,retrospective,inception,elicitation,research) - Extract
--limit=<n>if present (default: 5) - Everything else is the search query
- If no query provided, use recent conversation context
If query is empty:
- Extract key concepts from recent conversation (last 5-10 messages)
- Look for: file names, function names, error messages, technology terms
- Combine into a search query
Use Bash to invoke the Python library:
PLUGIN_ROOT="${CLAUDE_PLUGIN_ROOT:-$(ls -d ~/.claude/plugins/cache/git-notes-memory/memory-capture/*/ 2>/dev/null | head -1)}"
uv run --directory "$PLUGIN_ROOT" python3 -c "
from git_notes_memory import get_recall_service
recall = get_recall_service()
results = recall.search(
query='''$QUERY''',
namespace=$NAMESPACE, # None for all namespaces
k=$LIMIT,
)
if not results:
print('No relevant memories found.')
else:
print(f'## Recalled Memories ({len(results)} results)\n')
for i, r in enumerate(results, 1):
# Use summary (not title) and timestamp (not created_at)
print(f'### {i}. {r.namespace.title()}: {r.summary[:50]}')
print(f'**Relevance**: {r.score:.2f} | **Captured**: {r.timestamp.strftime(\"%Y-%m-%d\")}')
print(f'> {r.content[:200]}...\n')
"
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.
- yesterday First seen · 189 lines · 11 tokens per session scan A 41354b08f3c1
recall is a command published in the GitHub repository zircote/git-notes-memory (4 stars, last pushed 8mo ago), licensed MIT. It adds 11 tokens to every session and 1,328 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
memory-graph
Open the memory graph visualization in your browser.
memory-recall
Turn this project's automatic memory recall off or back on (storing stays active).
memory-status
Show a memory health digest for this project (items, sessions, injections served).
assets
Inventory .claude-design/ / — file counts, types, locations.
start
Run the canvas-to-code workflow — guided discovery, auto-resume, or jump to a specific gate.
dashboard
Work timeline — what's done, what's in progress, when, with PR + doc links.