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 hooks/jagoff/memo/session-startgit clone --depth 1 https://github.com/jagoff/memoGrade A, and why
SessionStart 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 2d ago.
A static scan of the commands this hook runs, not an audit. A hook is shell that executes on your machine at the event it names, which is why every command in it is printed with what was found.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
{
"SessionStart": [
{
"matcher": "startup|resume|clear",
"hooks": [
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 memo sync pull --quiet",
"timeout": 90,
"async": true,
"statusMessage": "memo: pulling cross-Mac memories…",
"comment": "Cross-Mac git pull + signal merge + reindex (F5). --quiet soft-fails (exit 0) when the memorias dir isn't a git clone, so non-synced installs are unaffected. Async so startup isn't blocked."
}
]
},
{
"matcher": "startup|clear",
"hooks": [
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 memo reflect --last --if-due --quiet",
"timeout": 90,
"async": true,
"comment": "Synthesize the previous session into durable memorias before briefing. --if-due skips if already done. Runs async so startup isn't blocked."
},
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 memo prewarm",
"timeout": 30,
"async": true,
"statusMessage": "memo: warming MLX models…"
},
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 memo recall-daemon start",
"timeout": 5,
"async": true,
"statusMessage": "memo: starting recall daemon…"
},
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 memo session recent",
"timeout": 5
},
{
"type": "command",
"command": "MEMO_NONINTERACTIVE=1 MEMO_SYNTHESIS_ENABLED=1 memo maintain --if-due",
"timeout": 5,
"async": true,
"comment": "Daily corpus freshness (archive-only, reversible). No-op unless >24h since last run; spawns detached so it never blocks startup. Disable with MEMO_MAINTAIN_DISABLE=1."
}
]
},
{
"matcher": "startup|resume|compact",
"hooks": [
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.
- 2d ago First seen · 65 lines scan A 1ca362f7eeab
SessionStart is a hook published in the GitHub repository jagoff/memo (16 stars, last pushed 2d ago), licensed MIT. Its token cost is not measured: a hook is shell that never enters the context. 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 2026-08-30.
Other hooks, from other repositories
PostToolUse
Runs after a tool call finishes for Read, Edit, Write and NotebookEdit tool calls, executing post.sh. From T0mSIlver/localvoxtral.
Stop
Runs when the agent finishes a response, executing post.sh. From T0mSIlver/localvoxtral.
SessionEnd
Runs when a session ends, executing post.sh. From T0mSIlver/localvoxtral.
CwdChanged
Runs on the CwdChanged event, executing post.sh. From T0mSIlver/localvoxtral.
UserPromptSubmit
Runs when you submit a prompt, before the agent sees it, executing post.sh. From T0mSIlver/localvoxtral.
PermissionRequest
Runs when the agent asks permission for an action for Bash tool calls, executing run-permission-notify.sh and approve-tts.py (2 commands). From aperepel/claude-mlx-tts.