agent-gauntlet 0.15.3 → 0.15.4

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (4) hide show
  1. package/dist/index.js +130 -84
  2. package/dist/index.js.map +9 -8
  3. package/package.json +1 -1
  4. package/skills/gauntlet-run/SKILL.md +2 -2
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "agent-gauntlet",
3
- "version": "0.15.3",
3
+ "version": "0.15.4",
4
4
  "description": "A CLI tool for testing AI coding agents",
5
5
  "license": "MIT",
6
6
  "author": "Paul Caplan",
package/skills/gauntlet-run/SKILL.md CHANGED
@@ -35,7 +35,7 @@ Execute the autonomous verification suite.
35
35
  ## Procedure
36
36
 
37
37
  1. Run `agent-gauntlet clean` to archive any previous log files.
38
- 2. Run `agent-gauntlet run` using `Bash` with `timeout: 300000`. Wait for the complete output. **Verify you can see a `Status:` line in the output before continuing.**
38
+ 2. Run `agent-gauntlet run 2>&1` using `Bash` with `timeout: 300000`. Wait for the complete output. **Verify you can see a `Status:` line in the output before continuing.**
39
39
  3. **Check the status line:**
40
40
  - `Status: Passed` → Go to step 8.
41
41
  - `Status: Passed with warnings` → Go to step 8.
@@ -59,7 +59,7 @@ Execute the autonomous verification suite.
59
59
  a. **Task tool** (Claude Code): `Task` with `subagent_type="general-purpose"`, `model="haiku"`, `prompt=` update-prompt content + log directory + decisions list. NEVER use `run_in_background: true`.
60
60
  b. **Subagent delegation**: Delegate the update-prompt instructions with the log directory and decisions to a subagent.
61
61
  c. **Inline fallback**: Follow the update-prompt instructions yourself to update the review JSON files.
62
- 7. **Re-run verification:** Run `agent-gauntlet run` again with `Bash` and `timeout: 300000`. Do NOT run `agent-gauntlet clean` between retries. The tool detects existing logs and automatically switches to verification mode. **Go back to step 3** to check the status line and repeat.
62
+ 7. **Re-run verification:** Run `agent-gauntlet run 2>&1` again with `Bash` and `timeout: 300000`. Do NOT run `agent-gauntlet clean` between retries. The tool detects existing logs and automatically switches to verification mode. **Go back to step 3** to check the status line and repeat.
63
63
  8. **Provide a summary** of the session:
64
64
  - Final Status: (Passed / Passed with warnings / Retry limit exceeded)
65
65
  - Issues Fixed: (list key fixes)