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 skills/soleimanmansouri/debug-bank/debug-trajectorynpx skills add soleimanmansouri/debug-bank --skill debug-trajectorygit clone --depth 1 https://github.com/soleimanmansouri/debug-bankWhat 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 | $0.00037 | $0.00745 |
| Opus 5 | $0.00018 | $0.00373 |
| Sonnet 5 | $0.00007 | $0.00149 |
| Haiku 4.5 | $0.00004 | $0.00075 |
Grade A, and why
debug-trajectory 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debug Trajectory Protocol
You are now in debug trajectory mode. Follow this protocol exactly.
Step 1: Pattern Check (DO THIS FIRST)
Before investigating, scan the pattern bank for a matching root cause type.
Read all files in the patterns/ directory (or your project's pattern bank location). For each pattern, check if its symptoms match the current bug.
If a pattern matches:
- State which pattern (e.g., "This matches P08: Config Resolution Chain Gap")
- Read the full pattern file
- Apply the check list
- If the check list confirms, apply the fix strategy directly
- Skip to Step 6 (Fix) — no need for Steps 2-5
If no pattern matches:
- State "No pattern match" and proceed to Step 2
Step 2: Reproduce
Get the exact error. Capture FULL output — never summarize.
- Run the failing operation
- Capture complete logs, stack traces, HTTP responses
- Include 50+ lines of surrounding context
- Record exact reproduction steps
If you can't reproduce, add temporary logging and try again. Do NOT proceed without reproduction.
Step 3: Hypothesize
State 2-3 possible root causes, ranked by likelihood.
For each hypothesis:
- "If [hypothesis] is correct, then [specific test] would show [specific result]"
Step 4: Isolate
Test hypotheses ONE AT A TIME. Binary search.
- Start with the most likely hypothesis
- Disable/enable components to narrow scope
- If 3 hypotheses fail → TRIGGER 3-EXCHANGE RULE (see below)
Step 5: Diagnose
Identify the SINGLE root cause.
- Trace the full call chain: trigger → propagation → symptom
- Check for "root cause behind the root cause"
- Verify this cause explains ALL symptoms
Step 6: Fix
Minimal change addressing the root cause.
- Change as little as possible
- Verify ALL symptoms resolved
- Check for regressions
Step 7: Record
Add trajectory to your domain catalog:
### [Category] Short Title (YYYY-MM-DD)
- **Symptom:** What was observed
- **Root cause:** The actual cause
- **Fix:** What was changed
- **Key insight:** Generalizable lesson
- **Pattern:** P-number or "New"
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 · 104 lines · 37 tokens per session scan A 12b128937047
debug-trajectory is a skill published in the GitHub repository soleimanmansouri/debug-bank (1 stars, last pushed 3mo ago), licensed MIT. It adds 37 tokens to every session and 745 once invoked, about $0.0002 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 skills, from other repositories
weave
Parallel strand orchestration — decompose a task into 3+ independently scoped strands, fan out real subagents (worktree-isolated when they write files), and coordinate through a session file and Insight Capsules. Use ONLY when the user explicitly invokes it by name or slash command.
coding-agents-farm
To orchestrate parallel coding-agent farms (Claude, Codex, Copilot, Gemini, etc.) on isolated git worktrees.
mcp-media-inbox-triage
Find media, partnership, and high-intent requests buried in a shared Gmail inbox, summarize the useful ones, and route them to the right owner through Zapier MCP.
mcp-sheet-cache-bridge
Keep Zapier Tables or another MCP-connected record store as the source of truth while Google Sheets-based tools read an event-driven cache tab.
customer-deck-builder
Build tailored customer-facing sales or customer success decks from approved account context and generate the presentation with Gamma through Zapier MCP.
premortem
First-principles premortem on a high-stakes plan: assume failure 6–18 months out, reconstruct causal death chains, surface hidden assumptions, deliver a revised reversibility-weighted plan. Use ONLY when the user explicitly invokes it by name or slash command.