agenttrace-session-audit

1# agenttrace Session Audit
2 
3Use this skill when session logs need an operational read: spend, token burn, cache use, tool failures, retry loops, latency, health, anomalies, and CI gate readiness.
4 
5## Workflow
6 
71. Prefer the installed `agenttrace` binary when it is available on `PATH`.
82. If the binary is not available and the current directory is the `luoyuctl/agenttrace` repository, use `cargo run -q -p agenttrace --`.
93. Start with discovery unless the user gave a specific file or directory:
10 
11```bash
12agenttrace --doctor
13agenttrace --overview
14```
15 
164. For a fast human report, use Markdown:
17 
18```bash
19agenttrace --overview -f markdown -o agenttrace-overview.md
20```
21 
225. For automation or CI, use JSON or health gates:
23 
24```bash
25agenttrace --overview -f json -o agenttrace-overview.json
26agenttrace --overview --fail-under-health 80 --fail-on-critical --max-tool-fail-rate 15
27```
28 
296. For a single recent session:
30 
31```bash
32agenttrace --latest
33agenttrace --latest -f json
34```
35 
367. For a specific export or session directory:
37 
38```bash
39agenttrace path/to/session-or-export.json
40agenttrace --overview -d path/to/session-dir
41```
42 
43## Report Focus
44 
45- Lead with the highest-risk sessions and the reason they matter.
46- Call out token/cost waste, repeated tool failures, retry loops, long gaps, and low health scores.
47- When proposing a CI gate, include the exact `agenttrace` command and threshold.
48…