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/assafkip/kipi-system/post-tool-usegit clone --depth 1 https://github.com/assafkip/kipi-systemGrade A, and why
PostToolUse 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 today.
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.
Copies of this mod
1 near-identical copy found in the catalogue:
- Stop — 86% identical, 127 lines differ
How it starts
The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
{
"PostToolUse": [
{
"matcher": "Edit|Write|MultiEdit|Bash",
"hooks": [
{
"type": "command",
"command": "python3 \"$CLAUDE_PROJECT_DIR/q-system/.q-system/token-guard.py\"",
"timeout": 5
},
{
"type": "command",
"command": "for R in q-system/.q-system/scripts/claude-path-write-guard.py q-system/.q-system/scripts/claude-integrity-tripwire.py; do [ -f \"$CLAUDE_PROJECT_DIR/$R\" ] && continue; git -C \"$CLAUDE_PROJECT_DIR\" checkout -- \"$R\" >/dev/null 2>&1; N=\"${KIPI_NOTIFY:-$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/slack-notify.sh}\"; if [ -f \"$CLAUDE_PROJECT_DIR/$R\" ]; then \"$N\" \"SECURITY: $R was deleted; restored from git\" >/dev/null 2>&1; else \"$N\" \"SECURITY: $R was deleted and git could not restore it -- that layer is DOWN\" >/dev/null 2>&1; echo \"SECURITY: $R is missing and could not be restored\" >&2; exit 2; fi; done",
"timeout": 15
},
{
"type": "command",
"command": "test -f \"$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/claude-integrity-tripwire.py\" && python3 \"$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/claude-integrity-tripwire.py\" --enforce --quiet",
"timeout": 20
}
]
},
{
"matcher": "Edit|Write|MultiEdit",
"hooks": [
{
"type": "command",
"command": "python3 \"$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/prompt-only-enforcement-guard.py\"",
"timeout": 5
},
{
"type": "command",
"command": "python3 \"$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/enforced-claim-lint.py\"",
"timeout": 5
},
{
"type": "command",
"command": "python3 \"$CLAUDE_PROJECT_DIR/q-system/.q-system/scripts/wiring-check.py\"",
"timeout": 5
},
{
"type": "command",
"command": "test -f \"$CLAUDE_PROJECT_DIR/q-system/.q-systeWhat 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.
- today Changed · +15 lines 6ee0954de191
- yesterday First seen · 142 lines scan A fad862a15b19
PostToolUse is a hook published in the GitHub repository assafkip/kipi-system (108 stars, last pushed today), 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
UserPromptSubmit
Runs when you submit a prompt, before the agent sees it, executing test-wrapper.sh. From modeled-information-format/mnemonic.
Stop
Runs when the agent finishes a response, executing capture.mjs via node. From aayoawoyemi/Ori-Mnemos.
TaskCompleted
Runs when a task is marked completed, running npx hipocampus compact. From kevin-hs-sohn/hipocampus.
SessionStart
Runs when a session starts, executing session_start.js via node. From remember-md/remember.
UserPromptSubmit
Runs when you submit a prompt, before the agent sees it, executing user_prompt.js via node. From remember-md/remember.
SessionStart
Runs when a session starts on startup and resume, running an inline shell check (2 commands). From sublimecoder/aios.