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/jmanhype/speckit/pre-compactgit clone --depth 1 https://github.com/jmanhype/speckitGrade A, and why
PreCompact 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.
This is a copy
89% identical to SessionStart — 2 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
{
"PreCompact": [
{
"hooks": [
{
"command": "command -v bd >/dev/null 2>&1 && bd prime || true",
"type": "command"
}
],
"matcher": ""
}
]
}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.
- today First seen · 13 lines scan A 8072374b611c
PreCompact is a hook published in the GitHub repository jmanhype/speckit (26 stars, last pushed 5mo 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. It is 89% identical to SessionStart, differing in 2 lines, and is treated as a copy.
Other hooks, from other repositories
PostToolUse
Runs after a tool call finishes for Bash tool calls, running an inline shell check. From zigrivers/scaffold.
PreToolUse
Runs before the agent uses a tool for Bash, Write and Write tool calls, running an inline shell check (3 commands). From Justdvp/claude-code-templates.
UserPromptSubmit
Runs when you submit a prompt, before the agent sees it, executing route-natural-language-aliases.sh. From loiane/specs-driven-development-spring-angular.
PreToolUse
Runs before the agent uses a tool for Bash tool calls, executing git-guard.mjs via node. From crenshawdev/cadence.
SessionStart
Runs when a session starts on startup, clear and compact, running an inline shell check. From voidvec/drogon-claude-plugin.
PreToolUse
Runs before the agent uses a tool for Bash, Read, Write, Edit, Glob, Grep, mcp__.*, Bash, Task, Write and Edit tool calls, executing dispatch.cjs, pre-dispatch-conflict-check.cjs, mcp-call-gate.cjs, commit-attribution.cjs, task-track-model.cjs, security-critical-warn.cjs and phase-checkpoint.cjs via node (8 commands).…