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/ogrodev/fsociety/debriefgit clone --depth 1 https://github.com/ogrodev/fsocietyWhat 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 | $0.00010 | $0.00995 |
| Opus 5 | $0.00005 | $0.00498 |
| Sonnet 5 | $0.00002 | $0.00199 |
| Haiku 4.5 | $0.00001 | $0.00100 |
Grade A, and why
debrief scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Storage Policy: ALL output files MUST be saved in the project directory. NEVER write to
/tmp/or any system temporary directory.
Post-Engagement Debrief
Analyze the entire engagement and produce a structured lessons-learned report.
Phase 1: Gather All Data
Run these in parallel:
node "${CLAUDE_PLUGIN_ROOT}/scripts/findings-tracker.js" list
node "${CLAUDE_PLUGIN_ROOT}/scripts/findings-tracker.js" summary
node "${CLAUDE_PLUGIN_ROOT}/scripts/techniques-tracker.js" list
node "${CLAUDE_PLUGIN_ROOT}/scripts/techniques-tracker.js" summary
node "${CLAUDE_PLUGIN_ROOT}/scripts/techniques-tracker.js" coverage
node "${CLAUDE_PLUGIN_ROOT}/scripts/chain-detector.js"
node "${CLAUDE_PLUGIN_ROOT}/scripts/session-state.js" show
node "${CLAUDE_PLUGIN_ROOT}/scripts/scan-profile.js" show 2>/dev/null
node "${CLAUDE_PLUGIN_ROOT}/scripts/target-intel.js" summary 2>/dev/null
Also check for and read:
scope.md— original engagement scopeengagement-context.md— engagement rules- Any
plan-*.mdorPLAN.mdfiles — original attack plans attacks/WEAPONIZATION-REPORT.md— if weaponization was done
Phase 2: Analyze
Build analysis across these dimensions:
What Worked
- Techniques with
successorpartialresults - Findings by severity — what was the highest-impact discovery?
- Chains detected — which combinations proved most dangerous?
- Tools with highest success rates
What Was Blocked
- Techniques with
blockedresult — what defenses stopped us? - Active blockers from session state — were any unresolved?
- Profile escalations — did we need to switch from loud to stealth?
Coverage Gaps
- From techniques coverage report — what categories were never tested?
- Attack surface not reached — endpoints, services, protocols untouched
- Tools available but never used
Effort Distribution
- How many techniques per target?
- How many sessions were needed?
- Time from first scan to final finding
- Scans-to-findings ratio (efficiency)
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 · 131 lines · 10 tokens per session scan A 88e4fccf5764
debrief is a command published in the GitHub repository ogrodev/fsociety (20 stars, last pushed 5mo ago), licensed MIT. It adds 10 tokens to every session and 995 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 2026-08-30.
Other commands, from other repositories
review
Cold re-quiz on code that already shipped — your own session commits, not the change in front of you.
no-vibe
Enter no-vibe mode in OpenCode (tutor mode, no direct project file writes).
OPSX: Onboard
Guided onboarding - walk through a complete OpenSpec workflow cycle with narration.
start-3-2-2
Module 3.2.2: Strategy & Architecture Visuals - Create system architecture diagrams, prioritization matrices, and product roadmaps.
recall
Search the engineering memlog and inject the top matching prior lessons. Use when you suspect a recurring issue.
discuss
Capture user preferences and lock decisions BEFORE planning begins. Prevents replanning from misaligned assumptions.