heimdall-agent 1.0.0 → 1.1.0

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 (231) hide show
  1. package/README.md +57 -26
  2. package/README.pt-BR.md +429 -0
  3. package/agents/cloud-engineer.md +77 -0
  4. package/agents/code-sentinel.md +66 -35
  5. package/agents/database-architect.md +31 -25
  6. package/agents/frontend-designer.md +30 -28
  7. package/agents/mobile-designer.md +25 -24
  8. package/agents/modernizer.md +66 -0
  9. package/agents/orchestrator.md +94 -93
  10. package/agents/reviewer.md +100 -71
  11. package/agents/security-auditor.md +27 -27
  12. package/agents/tester.md +97 -90
  13. package/commands/grill-me.md +20 -4
  14. package/commands/orchestrate.md +16 -0
  15. package/commands/orchestrator.md +16 -0
  16. package/commands/plan.md +17 -1
  17. package/config/agents.json +21 -3
  18. package/config/mcp.json +72 -0
  19. package/config/models.json +26 -16
  20. package/config/profiles.json +2 -4
  21. package/config/skills.json +210 -100
  22. package/docs/SKILLS.md +341 -245
  23. package/docs/designs/old/2026-10-31 plan-lifecycle-backlog-and-clean-code-skill.md +87 -0
  24. package/docs/designs/old/2026-11-02 skill-upstream-audit-and-update.md +105 -0
  25. package/docs/designs/old/2026-11-05 communication-doctrine-bluf-ste.md +94 -0
  26. package/docs/designs/old/2026-11-06-cloud-engineer-and-resilience-synthesis.md +67 -0
  27. package/docs/designs/old/2026-11-06-herdr-native-agent-orchestration.md +45 -0
  28. package/docs/runs/skill-audit/add_progressive_disclosure.cjs +30 -0
  29. package/docs/runs/skill-audit/check_heimdall_skills.ps1 +9 -0
  30. package/docs/runs/skill-audit/compare_classics.ps1 +26 -0
  31. package/docs/runs/skill-audit/compare_taste.ps1 +12 -0
  32. package/docs/runs/skill-audit/customization-report.json +380 -0
  33. package/docs/runs/skill-audit/diff_pending_skills.cjs +94 -0
  34. package/docs/runs/skill-audit/fetch_and_apply_skill.cjs +58 -0
  35. package/docs/runs/skill-audit/fetch_and_apply_taste_v2.cjs +65 -0
  36. package/docs/runs/skill-audit/inspect_pending_skills.cjs +71 -0
  37. package/docs/runs/skill-audit/inventory.json +308 -0
  38. package/docs/runs/skill-audit/list_local_taste.ps1 +1 -0
  39. package/docs/runs/skill-audit/phase5-update-plan.json +595 -0
  40. package/docs/runs/skill-audit/phase5-update-plan.md +138 -0
  41. package/docs/runs/skill-audit/probe.ps1 +16 -0
  42. package/docs/runs/skill-audit/probe_classics_branches.ps1 +3 -0
  43. package/docs/runs/skill-audit/probe_commits.ps1 +15 -0
  44. package/docs/runs/skill-audit/probe_repos.ps1 +17 -0
  45. package/docs/runs/skill-audit/probe_sentry_sec.ps1 +3 -0
  46. package/docs/runs/skill-audit/probe_sentry_sweep.ps1 +5 -0
  47. package/docs/runs/skill-audit/probe_snyk.ps1 +3 -0
  48. package/docs/runs/skill-audit/probe_snyk_recipes.ps1 +5 -0
  49. package/docs/runs/skill-audit/probe_taste_branches.ps1 +8 -0
  50. package/docs/runs/skill-audit/probe_taste_v19.ps1 +3 -0
  51. package/docs/runs/skill-audit/probe_user_repos.ps1 +19 -0
  52. package/docs/runs/skill-audit/prune_taste_skills.cjs +22 -0
  53. package/docs/runs/skill-audit/run_phase1_2.cjs +142 -0
  54. package/docs/runs/skill-audit/run_phase6_verification.cjs +107 -0
  55. package/docs/runs/skill-audit/show_16_details.ps1 +25 -0
  56. package/docs/runs/skill-audit/show_20_skills.ps1 +28 -0
  57. package/docs/runs/skill-audit/show_classics_diff.cjs +37 -0
  58. package/docs/runs/skill-audit/sync_agent_markdowns.cjs +30 -0
  59. package/docs/runs/skill-audit/taste-skill-upstream.md +542 -0
  60. package/docs/runs/skill-audit/update-manifest.md +88 -0
  61. package/docs/runs/skill-audit/update_config_skills.cjs +52 -0
  62. package/docs/runs/skill-audit/update_registry_phase7.cjs +83 -0
  63. package/docs/runs/skill-audit/upstream-status.json +563 -0
  64. package/docs/runs/skill-audit/verification-results.json +252 -0
  65. package/doctrine/AGENTS.md +39 -77
  66. package/hooks/git/pre-commit +11 -0
  67. package/hooks/heimdall-hook.cjs +81 -2
  68. package/hooks/herdr-agent-state.ps1 +55 -0
  69. package/hooks/hooks.json +0 -11
  70. package/lib/catalog-sync.cjs +162 -0
  71. package/lib/hermes-context.cjs +57 -0
  72. package/lib/model-context.cjs +123 -0
  73. package/lib/path-discovery.cjs +94 -0
  74. package/package.json +7 -3
  75. package/playbooks/01-interactive-grilling-gate.md +14 -1
  76. package/playbooks/02-fast-code-deep-verification.md +3 -2
  77. package/playbooks/03-adversarial-code-review.md +3 -3
  78. package/playbooks/05-canonical-plan-template.md +32 -18
  79. package/playbooks/06-model-registry-matrix.md +24 -13
  80. package/playbooks/07-legacy-code-modernization.md +74 -0
  81. package/playbooks/08-backlog-and-change-request-flow.md +122 -0
  82. package/setup.mjs +592 -172
  83. package/skills/antislop/SKILL.md +76 -56
  84. package/skills/antislop/VERSION +1 -0
  85. package/skills/antislop-copywriting/SKILL.md +372 -372
  86. package/skills/antislop-human/LICENSE-web-quality-skills +21 -0
  87. package/skills/antislop-human/SKILL.md +204 -147
  88. package/skills/antislop-human/contrast-check.py +168 -165
  89. package/skills/antislop-human/contrast-mcp.py +6 -3
  90. package/skills/antislop-human/references/A11Y-PATTERNS.md +249 -0
  91. package/skills/antislop-human/references/WCAG.md +127 -0
  92. package/skills/antislop-ui/LICENSE-web-quality-skills +21 -0
  93. package/skills/antislop-ui/SKILL.md +344 -304
  94. package/skills/antislop-ui/references/CORE-WEB-VITALS.md +181 -0
  95. package/skills/antislop-ui/references/WEB-BASELINE.md +28 -0
  96. package/skills/brainstorming/SKILL.md +274 -265
  97. package/skills/brainstorming/builder-check-prompt.md +23 -0
  98. package/skills/brainstorming/scripts/frame-template.html +213 -213
  99. package/skills/brainstorming/scripts/server.cjs +723 -723
  100. package/skills/brandkit/SKILL.md +798 -798
  101. package/skills/brutalist-skill/SKILL.md +93 -92
  102. package/skills/caveman/README.md +68 -0
  103. package/skills/caveman/SKILL.md +73 -38
  104. package/skills/claude-collaborator/SKILL.md +24 -44
  105. package/skills/clean-code/SKILL.md +152 -0
  106. package/skills/cloudflare-security-audit/AI-AND-LLM.md +83 -83
  107. package/skills/cloudflare-security-audit/ATTACK-CLASSES.md +130 -130
  108. package/skills/cloudflare-security-audit/CLIENT-SIDE.md +83 -83
  109. package/skills/cloudflare-security-audit/CLOUD-AND-DEPLOYMENT.md +86 -86
  110. package/skills/cloudflare-security-audit/DATA-ISOLATION-AND-LIFECYCLE.md +84 -84
  111. package/skills/cloudflare-security-audit/DESKTOP-MOBILE-AND-LOCAL-IPC.md +89 -89
  112. package/skills/cloudflare-security-audit/HUNTING.md +251 -251
  113. package/skills/cloudflare-security-audit/MEMORY-SAFETY-AND-BINARY.md +101 -101
  114. package/skills/cloudflare-security-audit/PROTOCOLS-RPC-AND-MESSAGING.md +81 -81
  115. package/skills/cloudflare-security-audit/RECONNAISSANCE.md +156 -156
  116. package/skills/cloudflare-security-audit/RESOURCE-EXHAUSTION-AND-AVAILABILITY.md +78 -78
  117. package/skills/cloudflare-security-audit/SKILL.md +192 -192
  118. package/skills/cloudflare-security-audit/SUPPLY-CHAIN-AND-RELEASE.md +73 -73
  119. package/skills/cloudflare-security-audit/VALIDATION-AND-REPORTING.md +186 -186
  120. package/skills/cloudflare-security-audit/WEB-PROTOCOL-AND-AUTH.md +105 -105
  121. package/skills/cloudflare-security-audit/report-schema.json +461 -461
  122. package/skills/cloudflare-security-audit/validate-coverage-ledger.cjs +872 -872
  123. package/skills/cloudflare-security-audit/validate-coverage-ledger.test.cjs +740 -740
  124. package/skills/cloudflare-security-audit/validate-findings.cjs +773 -773
  125. package/skills/cloudflare-security-audit/validate-findings.test.cjs +652 -652
  126. package/skills/codex-collaborator/SKILL.md +24 -45
  127. package/skills/communication-doctrine/SKILL.md +89 -0
  128. package/skills/diagnosing-superpowers/SKILL.md +1 -1
  129. package/skills/diagnosing-superpowers/references/github-issues.md +3 -3
  130. package/skills/diagnosing-superpowers/templates/issue.md +23 -4
  131. package/skills/executing-plans/SKILL.md +0 -52
  132. package/skills/executing-plans/scripts/task-done +52 -52
  133. package/skills/executing-plans/scripts/task-start +28 -28
  134. package/skills/gemini-collaborator/SKILL.md +24 -56
  135. package/skills/herdr/SKILL.md +243 -0
  136. package/skills/image-to-code-skill/SKILL.md +1228 -1228
  137. package/skills/imagegen-frontend-mobile/SKILL.md +1465 -1465
  138. package/skills/imagegen-frontend-web/SKILL.md +987 -987
  139. package/skills/karpathy-guidelines/SKILL.md +67 -67
  140. package/skills/mcp-engineering/SKILL.md +74 -0
  141. package/skills/minimalist-skill/SKILL.md +87 -85
  142. package/skills/output-skill/SKILL.md +49 -49
  143. package/skills/ponytail/SKILL.md +36 -52
  144. package/skills/postgres-best-practices/SKILL.md +30 -52
  145. package/skills/postgres-best-practices/references/backup-restore.md +372 -0
  146. package/skills/postgres-best-practices/references/bulk-loading.md +315 -0
  147. package/skills/postgres-best-practices/references/connection-pooling.md +310 -0
  148. package/skills/postgres-best-practices/references/hot-standby.md +221 -0
  149. package/skills/postgres-best-practices/references/indexing.md +318 -0
  150. package/skills/postgres-best-practices/references/logical-replication.md +610 -0
  151. package/skills/postgres-best-practices/references/major-version-upgrades.md +334 -0
  152. package/skills/postgres-best-practices/references/performance-diagnostics.md +553 -0
  153. package/skills/postgres-best-practices/references/query-optimization.md +294 -0
  154. package/skills/postgres-best-practices/references/query-patterns.md +566 -0
  155. package/skills/postgres-best-practices/references/schema-design.md +362 -0
  156. package/skills/postgres-best-practices/references/security-roles.md +489 -0
  157. package/skills/postgres-best-practices/references/transaction-isolation.md +344 -0
  158. package/skills/progressive-disclosure/SKILL.md +68 -0
  159. package/skills/pstack-discipline/SKILL.md +40 -0
  160. package/skills/receiving-code-review/SKILL.md +2 -23
  161. package/skills/requesting-code-review/SKILL.md +7 -36
  162. package/skills/soft-skill/SKILL.md +99 -98
  163. package/skills/stitch-skill/DESIGN.md +121 -121
  164. package/skills/stitch-skill/SKILL.md +185 -184
  165. package/skills/strix-security/SKILL.md +54 -53
  166. package/skills/subagent-driven-development/SKILL.md +28 -260
  167. package/skills/subagent-driven-development/references/batching.md +11 -0
  168. package/skills/subagent-driven-development/references/cannot-verify.md +10 -0
  169. package/skills/subagent-driven-development/references/final-review.md +47 -0
  170. package/skills/subagent-driven-development/references/fix-loop.md +77 -0
  171. package/skills/subagent-driven-development/references/handling-reports.md +15 -0
  172. package/skills/subagent-driven-development/references/preflight.md +26 -0
  173. package/skills/subagent-driven-development/references/resuming.md +12 -0
  174. package/skills/subagent-driven-development/scripts/review-package +53 -53
  175. package/skills/subagent-driven-development/scripts/sdd-workspace +87 -82
  176. package/skills/subagent-driven-development/scripts/task-brief +43 -43
  177. package/skills/systematic-debugging/SKILL.md +283 -306
  178. package/skills/taste-skill/SKILL.md +542 -1206
  179. package/skills/taste-skill/references/app-screens.md +1199 -0
  180. package/skills/taste-skill/references/colour.md +141 -0
  181. package/skills/taste-skill/references/copy.md +177 -0
  182. package/skills/taste-skill/references/critique.md +182 -0
  183. package/skills/taste-skill/references/imagery.md +791 -0
  184. package/skills/taste-skill/references/layout.md +292 -0
  185. package/skills/taste-skill/references/motion.md +980 -0
  186. package/skills/taste-skill/references/research.md +356 -0
  187. package/skills/taste-skill/references/scenes.md +174 -0
  188. package/skills/taste-skill/references/tells.md +115 -0
  189. package/skills/taste-skill/references/type.md +170 -0
  190. package/skills/test-driven-development/SKILL.md +330 -337
  191. package/skills/thermo/SKILL.md +67 -0
  192. package/skills/threat-modeling/SKILL.md +55 -0
  193. package/skills/trailofbits-audit-context/SKILL.md +72 -72
  194. package/skills/trailofbits-insecure-defaults/commands/audit.md +35 -0
  195. package/skills/trailofbits-insecure-defaults/references/debug-features.json +13 -13
  196. package/skills/trailofbits-insecure-defaults/references/debug-features.md +75 -75
  197. package/skills/trailofbits-insecure-defaults/references/default-credentials.json +12 -12
  198. package/skills/trailofbits-insecure-defaults/references/default-credentials.md +76 -76
  199. package/skills/trailofbits-insecure-defaults/references/fail-open-security.json +11 -11
  200. package/skills/trailofbits-insecure-defaults/references/fail-open-security.md +70 -70
  201. package/skills/trailofbits-insecure-defaults/references/fallback-secrets.json +11 -11
  202. package/skills/trailofbits-insecure-defaults/references/fallback-secrets.md +70 -70
  203. package/skills/trailofbits-insecure-defaults/references/permissive-access.json +13 -13
  204. package/skills/trailofbits-insecure-defaults/references/permissive-access.md +62 -62
  205. package/skills/trailofbits-insecure-defaults/references/weak-crypto.json +13 -13
  206. package/skills/trailofbits-insecure-defaults/references/weak-crypto.md +71 -71
  207. package/skills/trailofbits-triage-brocards/SKILL.md +208 -208
  208. package/skills/verification-before-completion/SKILL.md +0 -14
  209. package/skills/writing-plans/SKILL.md +11 -15
  210. package/skills/writing-skills/SKILL.md +21 -303
  211. package/skills/writing-skills/references/CLAUDE_MD_TESTING.md +189 -0
  212. package/skills/writing-skills/references/anthropic-best-practices.md +1150 -0
  213. package/skills/writing-skills/references/bulletproofing.md +80 -0
  214. package/skills/writing-skills/references/discovery.md +201 -0
  215. package/skills/writing-skills/references/flowcharts.md +37 -0
  216. package/skills/writing-skills/references/graphviz-conventions.dot +172 -0
  217. package/skills/writing-skills/references/persuasion-principles.md +187 -0
  218. package/skills/writing-skills/references/testing-skills-with-subagents.md +384 -0
  219. package/tools/{claude-bridge.ps1 → archive/claude-bridge.ps1} +10 -7
  220. package/tools/{codex-bridge.ps1 → archive/codex-bridge.ps1} +1 -1
  221. package/tools/{gemini-bridge.ps1 → archive/gemini-bridge.ps1} +12 -19
  222. package/skills/gpt-tasteskill/SKILL.md +0 -74
  223. package/skills/taste-skill-v1/SKILL.md +0 -226
  224. /package/tools/{claude-bridge.cmd → archive/claude-bridge.cmd} +0 -0
  225. /package/tools/{claude-bridge.sh → archive/claude-bridge.sh} +0 -0
  226. /package/tools/{codex-bridge.cmd → archive/codex-bridge.cmd} +0 -0
  227. /package/tools/{codex-bridge.sh → archive/codex-bridge.sh} +0 -0
  228. /package/tools/{codexcli-bridge.cmd → archive/codexcli-bridge.cmd} +0 -0
  229. /package/tools/{codexcli-bridge.ps1 → archive/codexcli-bridge.ps1} +0 -0
  230. /package/tools/{gemini-bridge.cmd → archive/gemini-bridge.cmd} +0 -0
  231. /package/tools/{gemini-bridge.sh → archive/gemini-bridge.sh} +0 -0
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: codex-collaborator
3
3
  description: >
4
- Coordinate with OpenAI Codex CLI (codex.exe) to cross-validate changes, request second-opinion
4
+ Coordinate with OpenAI Codex CLI (codex.exe) via Herdr to cross-validate changes, request second-opinion
5
5
  adversarial reviews on git diffs, or validate test and build executions.
6
6
  upstream: https://github.com/bepaldes/heimdall-agent
7
7
  version: 1.0.0
@@ -9,19 +9,7 @@ version: 1.0.0
9
9
 
10
10
  # Codex Collaborator Skill
11
11
 
12
- This skill enables Google Antigravity and Claude Code to orchestrate tasks with OpenAI Codex CLI (`codex.exe`), delegating complex architectural reasoning, diff reviews, and adversarial validation across the tri-directional bridge.
13
-
14
- ## Real-Time Visibility in Herdr (Right Split & Auto-Close)
15
-
16
- By default when running inside Herdr, all bridge invocations automatically:
17
- - **Split the calling screen to the right**: Creates a side-by-side split pane in your current Herdr window.
18
- - **Display a prominent request banner**:
19
- - Caller AI $\rightarrow$ Target Model (`gpt-6-astra`)
20
- - Action & Working Directory
21
- - **`>> WHAT IS BEING REQUESTED:`** A highlighted box showing the exact prompt/instructions.
22
- - Summary of modified files (`git diff --stat`) for code reviews.
23
- - **Stream Codex's live execution**: Real-time display of reasoning, tool calls, and output right in that split pane.
24
- - **Auto-close when complete**: Automatically closes the right pane after finishing and collapses your screen back to normal, returning the captured output to the caller.
12
+ This skill enables Google Antigravity and Claude Code to orchestrate tasks with OpenAI Codex CLI (`codex.exe`) directly through native Herdr multiplexing, delegating complex architectural reasoning, diff reviews, and adversarial validation.
25
13
 
26
14
  ## When to Activate
27
15
 
@@ -29,48 +17,39 @@ By default when running inside Herdr, all bridge invocations automatically:
29
17
  - In multi-agent plans requiring tri-directional cross-model validation (Claude $\rightarrow$ Codex, or Gemini $\rightarrow$ Codex).
30
18
  - When validating complex TypeScript types, database migrations, or edge-case logic with an alternative reasoning perspective.
31
19
 
32
- ## Available Actions via `tools/codex-bridge.ps1`
20
+ ## Orchestration Method: Native Herdr Agent (`herdr` skill)
33
21
 
34
- Run commands using the project PowerShell helper script:
22
+ When running inside Herdr (`$env:HERDR_ENV = "1"`), communicate directly using Herdr's native agent orchestration. This maintains a persistent session across turns, preserves context, respects interactive approvals (`blocked`), and handles bracketed paste:
35
23
 
36
- ### 1. Adversarial Code Review of Git Diff
37
- Splits the screen to the right, runs Codex CLI review, streams live, and auto-closes:
38
- ```powershell
39
- powershell -ExecutionPolicy Bypass -File tools/codex-bridge.ps1 review
24
+ ### 1. Fast Path: Re-use Active Idle Agent
25
+ Before splitting a pane, check if a Codex agent is already running and idle:
26
+ ```bash
27
+ herdr agent list
40
28
  ```
41
- To provide specific review instructions:
42
- ```powershell
43
- powershell -ExecutionPolicy Bypass -File tools/codex-bridge.ps1 review "Focus on security vulnerabilities and boundary conditions"
29
+ If an agent named `codex-reviewer` or `codex` is idle or done, prompt it directly:
30
+ ```bash
31
+ herdr agent prompt codex-reviewer "Review the latest changes." --wait --timeout 120000
44
32
  ```
45
33
 
46
- ### 2. Consult Codex for Architectural Second Opinion
47
- Sends an inquiry or architectural prompt directly to Codex (Astra) in a right-split pane:
48
- ```powershell
49
- powershell -ExecutionPolicy Bypass -File tools/codex-bridge.ps1 ask "How should we design idempotency keys for external webhook processing?"
34
+ ### 2. Spawning a New Codex Agent (Single-Execution Batch)
35
+ If no active Codex agent exists, batch pane creation and agent startup in a single command:
36
+ ```bash
37
+ PANE_ID=$(herdr pane split --current --direction right --cwd "$PWD" --no-focus | jq -r '.result.pane.pane_id')
38
+ herdr agent start codex-reviewer --kind codex --pane "$PANE_ID"
39
+ herdr agent prompt codex-reviewer "Review git diff for security regressions and boundary conditions." --wait --timeout 120000
50
40
  ```
51
41
 
52
- ### 3. Validate Test or Build Run
53
- Runs a test command and has Codex analyze output and errors in a right-split pane:
54
- ```powershell
55
- powershell -ExecutionPolicy Bypass -File tools/codex-bridge.ps1 validate "pnpm test"
42
+ ### 3. Read Findings into Context
43
+ ```bash
44
+ herdr agent read codex-reviewer --source recent-unwrapped --lines 120
56
45
  ```
57
46
 
58
- ### 4. Interactive Handoff to Codex in a Right Split Pane
59
- Launches a full interactive Codex session in a right split pane pre-loaded with a task:
60
- ```powershell
61
- powershell -ExecutionPolicy Bypass -File tools/codex-bridge.ps1 interactive "Let's investigate this edge case together"
47
+ ### 4. Interactive Collaboration
48
+ When an interactive session or clarification is needed, send keys or prompt interactively:
49
+ ```bash
50
+ herdr agent send-keys codex-reviewer enter
62
51
  ```
63
52
 
64
- ### Display & Control Flags:
65
- - *(default)*: Splits the current Herdr screen to the right, streams live, and auto-closes when done.
66
- - `-KeepOpen`: Keeps the split pane open after completion instead of auto-closing.
67
- - `-Tab`: Opens in a new Herdr tab instead of splitting the screen.
68
- - `-Window`: Forces a standalone OS console window instead of Herdr.
69
- - `-Headless`: Runs silently in the background (for CI/unattended scripts).
70
- - `-NoFocus`: Prevents switching focus to the right pane.
71
- - `-Model <name>`: Override model (e.g. `-Model gpt-6-astra`).
72
- - `-Effort <level>`: Reasoning effort (`low`, `medium`, `high`, `xhigh`).
73
-
74
53
  ## Guidelines for Collaborating Agents
75
54
 
76
55
  - **Synthesize, Don't Blindly Echo**: When Codex returns feedback, analyze its critiques against the project's requirements and design system.
@@ -0,0 +1,89 @@
1
+ ---
2
+ name: communication-doctrine
3
+ description: >
4
+ Bottom Line Up Front (BLUF) and Simplified Technical English (STE) communication doctrine.
5
+ Enforces zero-latency verdicts in reports and unambiguous, imperative step definitions in plans.
6
+ ---
7
+
8
+ # Communication Doctrine: BLUF & Simplified Technical English (STE)
9
+
10
+ This skill governs structured communication across Google Antigravity, Claude Code, and OpenAI Codex CLI. It eliminates AI throat-clearing, hidden conclusions, and linguistic ambiguity.
11
+
12
+ ---
13
+
14
+ ## 1. Macro Architecture: Bottom Line Up Front (BLUF)
15
+
16
+ Always state the decision, status, verdict, or core recommendation in the **first 1–2 sentences** before presenting context, analysis, or supporting steps.
17
+
18
+ ### Schema A: Subagent Scratchpad Handoff (`docs/runs/<task_id>/report.md`)
19
+ Every subagent report must open with this exact 3-line block:
20
+ ```markdown
21
+ ### Subagent BLUF
22
+ - **Status**: SUCCESS | FAILURE | ESCALATED
23
+ - **Blast Radius**: [Files modified, tests added/executed, zero dev servers spawned]
24
+ - **Deliverable**: [Path to diff, artifact, or commit hash]
25
+ ```
26
+
27
+ ### Schema B: Code Reviewer & Security Sentinel Verdict (`agents/reviewer.md`)
28
+ Every review or audit output must open with this exact verdict block:
29
+ ```markdown
30
+ ### Review Verdict: APPROVED | BLOCKED | REVISION_NEEDED
31
+ - **Blockers**: <count>
32
+ - **Warnings**: <count>
33
+ - **Nits**: <count>
34
+ - **Mandatory Action**: [1-sentence summary of immediate action required before merge, or "None"]
35
+ ```
36
+
37
+ ### Schema C: Implementation Plan Header (`docs/designs/`)
38
+ Every implementation plan must lead with:
39
+ ```markdown
40
+ ### Plan BLUF
41
+ - **Goal**: [1 sentence describing exactly what is built]
42
+ - **Architectural Approach**: [1-2 sentences on core mechanism]
43
+ - **Risk & Blast Radius**: [Low | Medium | High — state key failure modes]
44
+ - **Human Approval Gate**: [Required before Phase X | Autonomous non-destructive]
45
+ ```
46
+
47
+ ---
48
+
49
+ ## 2. Micro Syntax: Simplified Technical English (STE)
50
+
51
+ Derived from ASD-STE100 principles, STE eliminates ambiguous interpretation, reduces token waste, and guarantees deterministic execution by child subagents.
52
+
53
+ ### Rule 1: Imperative Verb First
54
+ Every procedural step must start with an active imperative verb:
55
+ * **Approved Verbs**: `Create`, `Modify`, `Assert`, `Verify`, `Delete`, `Run`, `Open`, `Extract`, `Replace`, `Benchmark`.
56
+ * **Prohibited**: Passive constructions (*"The file should be updated..."*), conversational requests (*"Let's go ahead and..."*), or hedged steps (*"Consider checking..."*).
57
+
58
+ ### Rule 2: Strict Sentence Length Limits
59
+ * **Procedural Steps**: Maximum **20 words** per step. Each step represents exactly one checkable operation.
60
+ * **Descriptive Text**: Maximum **25 words** per sentence.
61
+
62
+ ### Rule 3: Banned Ambiguous Qualifiers (The Zero-Vagueness Lexicon)
63
+ The following words mask missing specifications and are strictly prohibited in implementation steps and suggested fixes:
64
+
65
+ | Banned Word | Why Prohibited | STE Replacement Pattern |
66
+ | :--- | :--- | :--- |
67
+ | `ensure` | Does not specify the mechanism. | `Verify <condition> equals <expected>. If not, <action>.` |
68
+ | `properly` | Subjective without testable criteria. | `Assert <field> matches regex <pattern>.` |
69
+ | `appropriately` | Defers specification to executor. | State the exact configuration parameter or value. |
70
+ | `gracefully` | Vague error handling. | `Catch <ErrorType>, log warning, and return <FallbackValue>.` |
71
+ | `as needed` / `as appropriate` | Unbounded scope. | Explicitly list the conditions or omit the step. |
72
+ | `robustly` / `seamlessly` | Pure AI marketing filler. | Delete the word. |
73
+ | `etc.` | Executor cannot guess remaining items. | List all items exhaustively. |
74
+
75
+ ### Rule 4: Explicit Object Naming (No Floating Pronouns)
76
+ Do not use ambiguous pronouns (*"it"*, *"this"*, *"they"*, *"that"*) when referring to code symbols, files, or error states. Always repeat the specific noun:
77
+ * *Bad*: *"Read the config and update it so it does not fail when it is missing."*
78
+ * *STE*: *"Read `config.json`. If `port` key is missing, assign default port `8080`."*
79
+
80
+ ---
81
+
82
+ ## 3. Communication Checklist
83
+
84
+ Before finalizing any plan, subagent report, or review output:
85
+ - [ ] Verdict or conclusion is on Line 1 (BLUF).
86
+ - [ ] Procedural steps start with an approved imperative verb.
87
+ - [ ] No procedural step exceeds 20 words.
88
+ - [ ] No banned qualifiers (`ensure`, `properly`, `appropriately`, `gracefully`, `as needed`, `etc.`).
89
+ - [ ] All code targets and file paths are explicitly named.
@@ -100,7 +100,7 @@ yourself in step 3 and which findings to lead with in the verdict.
100
100
  partner either.
101
101
  - **Approval gates.** No archive before your partner has seen the scrub
102
102
  log and file list. No issue or comment before they approve the exact
103
- text.
103
+ text, and none with a placeholder summary.
104
104
  - **Intake before analysis.** Nothing in steps 2–7 starts until your
105
105
  partner has answered. If they are away, write the questions and stop.
106
106
  A statement you reconstructed for them is not an answer. An
@@ -36,11 +36,11 @@ labels land only for collaborators; the template footer still marks the
36
36
  issue as skill-filed. `gh` cannot attach files: give your partner the
37
37
  bundle path to attach through the browser after the issue exists.
38
38
 
39
- Without `gh`, hand over a prefilled link on the `diagnosis_report.md`
40
- template, which applies both labels for any reporter:
39
+ Without `gh`, hand over a prefilled link on the `bug_report.md` template,
40
+ which applies the `bug` label for any reporter:
41
41
 
42
42
  ```
43
- https://github.com/obra/superpowers/issues/new?template=diagnosis_report.md&title=<url-encoded title>&body=<url-encoded body>
43
+ https://github.com/obra/superpowers/issues/new?template=bug_report.md&title=<url-encoded title>&body=<url-encoded body>
44
44
  ```
45
45
 
46
46
  GitHub rejects URLs over about 8,000 characters; past that, send the link
@@ -2,6 +2,20 @@ Title: <skill or symptom>: <one-line observable> (<harness>)
2
2
 
3
3
  - [x] I searched existing issues and this is not a duplicate (searched: <query terms>; closest: <#n title, or "none">)
4
4
 
5
+ ## Who's filing this report? (required)
6
+
7
+ <model and harness of this session>, filing on behalf of <how your
8
+ partner wants to be named, or "my human partner">, who asked: "<their
9
+ request from this session that led to the diagnosis or the issue,
10
+ scrubbed>"
11
+
12
+ ## Human-authored summary (required, under 200 words)
13
+
14
+ <Your partner's own words on what went wrong and what should have
15
+ happened, pasted as written. Ask for them; a placeholder here means the
16
+ issue is not ready to file. If they run long, ask them to trim rather
17
+ than editing their words.>
18
+
5
19
  ## Environment (required)
6
20
 
7
21
  | Field | Value | Provenance / supporting evidence |
@@ -17,8 +31,12 @@ Title: <skill or symptom>: <one-line observable> (<harness>)
17
31
 
18
32
  - [ ] I confirmed this issue does not occur without Superpowers installed
19
33
 
20
- The reporter has not tried reproducing without superpowers. Evidence for
21
- involvement is below; it does not establish cause.
34
+ How I verified that this issue does not reproduce without Superpowers:
35
+ <"Not tried", or what your partner ran without superpowers and what
36
+ happened, with the transcript path if there is one. Tick the box above
37
+ only when a transcript shows the problem did not recur.>
38
+
39
+ Evidence for involvement is below; it does not establish cause.
22
40
 
23
41
  ## What happened?
24
42
 
@@ -47,5 +65,6 @@ Superpowers involvement per the diagnosis report: <possible | likely>, with
47
65
  evidence at <transcript lines>. This report does not propose a fix.
48
66
 
49
67
  ---
50
- Filed with the `diagnosing-superpowers` skill. Model, harness, harness
51
- version, and installed plugins are listed above.
68
+ Filed with the `diagnosing-superpowers` skill. The filing model and
69
+ harness are under "Who's filing"; the diagnosed session's model, harness,
70
+ versions, and plugins are in the Environment table.
@@ -319,55 +319,3 @@ Use superpowers:finishing-a-development-branch.
319
319
  | "The reviewer said Minor, so it's Minor" | The label graded the spec's silence. Grade what the person gets. Re-grade, then gate. |
320
320
  | "The fix is obvious, no need for a failing test first" | The failing test is the only proof the finding was real and is now gone. Without it you have a diff and a hope. |
321
321
  | "I'll fix the minors too while I'm in there" | Every minor you fix is a test, a fix, and a suite run your partner did not ask for. Ledger them; your partner decides. |
322
-
323
- ## Example Workflow
324
-
325
- ```
326
- You: I'm using the executing-plans skill to implement this plan inline.
327
-
328
- [Setup: worktree verified]
329
- [Read plan once: docs/superpowers/plans/feature-plan.md; spec read]
330
- [Resolve workspace: sdd-workspace docs/superpowers/plans/feature-plan.md — no ledger inside, fresh start]
331
- [Pre-flight scan: 2 shared-interface rows, 4 self-consistency rows, clean; written to ledger]
332
- [Create todos for all tasks]
333
-
334
- Task 1: Hook installation script
335
-
336
- [task-start plan 1 → brief read; BASE a1b2c3d]
337
- [Step 1: write failing test — written]
338
- [Step 2: run it — FAIL: install_hook not defined. Matches Expected.]
339
- [Step 3: implement — written]
340
- [Step 4: run it — PASS 1/1. Matches Expected.]
341
- [Step 5: commit — d4e5f6a]
342
- [Contract: tests ran, output read, no deviations]
343
- [task-done plan 1 a1b2c3d -- npm test -- hooks → ledger: Task 1: complete (commits a1b2c3d..d4e5f6a, tests: npm test -- hooks → 1/1 pass)]
344
-
345
- Task 2: Recovery modes
346
-
347
- [task-start plan 2 → brief read; BASE d4e5f6a]
348
- [Step 2: run failing test — FAIL, but on an import error: Task 1 exported
349
- installHook, brief consumes install_hook]
350
- [Ruling: brief's consumer name is a typo against Task 1's Produces block;
351
- use installHook — Ledger: Task 2: Ruling: install_hook → installHook — matches Task 1 Produces — cost if wrong: one rename]
352
- [Steps 2-5 as planned; commit b7c8d9e]
353
- [task-done plan 2 d4e5f6a -- npm test -- recovery → ledger: Task 2: complete (commits d4e5f6a..b7c8d9e, tests: npm test -- recovery → 8/8 pass)]
354
-
355
- ...
356
-
357
- [After all tasks: review-package plan MERGE_BASE HEAD; dispatch code-reviewer, most capable model]
358
- Reviewer: One Important finding — progress reporting interval hardcoded. Two Minor.
359
- [Re-grade: Important stands; minors → ledger as deferred]
360
- [Fix pass: test_progress_interval_configurable RED → extract PROGRESS_INTERVAL → GREEN; suite 12/12; commit]
361
- [Ledger: Final: fixed hardcoded interval — test_progress_interval_configurable RED→GREEN, suite 12/12]
362
-
363
- Rulings I made:
364
- - Task 2: install_hook → installHook (brief typo; cost if wrong: one rename)
365
-
366
- Deferred minors:
367
- - README lacks a usage example
368
- - recovery.js could split verify/repair into two files
369
-
370
- [Delete this plan's workspace — the record now lives in git]
371
-
372
- Using superpowers:finishing-a-development-branch.
373
- ```
@@ -1,52 +1,52 @@
1
- #!/usr/bin/env bash
2
- # Close one task of an inline plan execution in a single call: run the task's
3
- # test command, keep its full output in the workspace, print the tail, and —
4
- # only if the command succeeded — append the completion line to the ledger.
5
- # A failing command records nothing: the task is not complete.
6
- #
7
- # Usage: task-done PLAN_FILE TASK_NUMBER BASE -- TEST_COMMAND [ARGS...]
8
- # BASE is the SHA task-start printed; the completion line records BASE..HEAD.
9
- # Exit: the test command's exit status.
10
- set -euo pipefail
11
-
12
- if [ $# -lt 5 ] || [ "$4" != "--" ]; then
13
- echo "usage: task-done PLAN_FILE TASK_NUMBER BASE -- TEST_COMMAND [ARGS...]" >&2
14
- exit 2
15
- fi
16
-
17
- plan=$1
18
- n=$2
19
- base=$3
20
- shift 4
21
- sdd="$(cd "$(dirname "$0")/../../subagent-driven-development/scripts" && pwd)"
22
-
23
- git rev-parse --verify --quiet "$base" >/dev/null || { echo "bad BASE: $base" >&2; exit 2; }
24
-
25
- dir=$("$sdd/sdd-workspace" "$plan")
26
- log="$dir/task-${n}-tests.log"
27
- ledger="$dir/progress.md"
28
-
29
- # Render the command the way a person would type it, for the ledger line.
30
- cmd=""
31
- for a in "$@"; do
32
- case "$a" in
33
- *[[:space:]\"\;\|\&]*) cmd="$cmd '$a'" ;;
34
- *) cmd="$cmd $a" ;;
35
- esac
36
- done
37
- cmd=${cmd# }
38
-
39
- rc=0
40
- "$@" > "$log" 2>&1 || rc=$?
41
-
42
- tail -n 5 "$log"
43
- if [ "$rc" -ne 0 ]; then
44
- echo "task-done: test command exited $rc; Task $n NOT recorded (full output: $log)" >&2
45
- exit "$rc"
46
- fi
47
-
48
- last=$(grep -v '^[[:space:]]*$' "$log" | tail -n 1)
49
- [ -f "$ledger" ] || printf '# SDD ledger — plan: %s\n' "$plan" > "$ledger"
50
- line="Task $n: complete (commits $(git rev-parse --short=7 "$base")..$(git rev-parse --short=7 HEAD), tests: $cmd → $last)"
51
- printf '%s\n' "$line" >> "$ledger"
52
- echo "ledger: $line"
1
+ #!/usr/bin/env bash
2
+ # Close one task of an inline plan execution in a single call: run the task's
3
+ # test command, keep its full output in the workspace, print the tail, and —
4
+ # only if the command succeeded — append the completion line to the ledger.
5
+ # A failing command records nothing: the task is not complete.
6
+ #
7
+ # Usage: task-done PLAN_FILE TASK_NUMBER BASE -- TEST_COMMAND [ARGS...]
8
+ # BASE is the SHA task-start printed; the completion line records BASE..HEAD.
9
+ # Exit: the test command's exit status.
10
+ set -euo pipefail
11
+
12
+ if [ $# -lt 5 ] || [ "$4" != "--" ]; then
13
+ echo "usage: task-done PLAN_FILE TASK_NUMBER BASE -- TEST_COMMAND [ARGS...]" >&2
14
+ exit 2
15
+ fi
16
+
17
+ plan=$1
18
+ n=$2
19
+ base=$3
20
+ shift 4
21
+ sdd="$(cd "$(dirname "$0")/../../subagent-driven-development/scripts" && pwd)"
22
+
23
+ git rev-parse --verify --quiet "$base" >/dev/null || { echo "bad BASE: $base" >&2; exit 2; }
24
+
25
+ dir=$("$sdd/sdd-workspace" "$plan")
26
+ log="$dir/task-${n}-tests.log"
27
+ ledger="$dir/progress.md"
28
+
29
+ # Render the command the way a person would type it, for the ledger line.
30
+ cmd=""
31
+ for a in "$@"; do
32
+ case "$a" in
33
+ *[[:space:]\"\;\|\&]*) cmd="$cmd '$a'" ;;
34
+ *) cmd="$cmd $a" ;;
35
+ esac
36
+ done
37
+ cmd=${cmd# }
38
+
39
+ rc=0
40
+ "$@" > "$log" 2>&1 || rc=$?
41
+
42
+ tail -n 5 "$log"
43
+ if [ "$rc" -ne 0 ]; then
44
+ echo "task-done: test command exited $rc; Task $n NOT recorded (full output: $log)" >&2
45
+ exit "$rc"
46
+ fi
47
+
48
+ last=$(grep -v '^[[:space:]]*$' "$log" | tail -n 1) || last="(no output)"
49
+ [ -f "$ledger" ] || printf '# SDD ledger — plan: %s\n' "$plan" > "$ledger"
50
+ line="Task $n: complete (commits $(git rev-parse --short=7 "$base")..$(git rev-parse --short=7 HEAD), tests: $cmd → $last)"
51
+ printf '%s\n' "$line" >> "$ledger"
52
+ echo "ledger: $line"
@@ -1,28 +1,28 @@
1
- #!/usr/bin/env bash
2
- # Begin one task of an inline plan execution in a single call: extract the
3
- # task's brief (via subagent-driven-development's task-brief, so both skills
4
- # share one workspace) and record BASE, the commit the task's review range is
5
- # cut from. One tool call instead of two, because every call in an inline
6
- # session is a turn that re-reads the whole context.
7
- #
8
- # Usage: task-start PLAN_FILE TASK_NUMBER
9
- # Prints:
10
- # brief: <path to the task's brief file>
11
- # base: <full SHA of HEAD>
12
- set -euo pipefail
13
-
14
- if [ $# -ne 2 ]; then
15
- echo "usage: task-start PLAN_FILE TASK_NUMBER" >&2
16
- exit 2
17
- fi
18
-
19
- plan=$1
20
- n=$2
21
- sdd="$(cd "$(dirname "$0")/../../subagent-driven-development/scripts" && pwd)"
22
-
23
- out=$("$sdd/task-brief" "$plan" "$n")
24
- brief=$(printf '%s\n' "$out" | sed -n 's/^wrote \(.*\): [0-9][0-9]* lines$/\1/p')
25
- [ -n "$brief" ] || { echo "task-brief did not report a path: $out" >&2; exit 1; }
26
-
27
- echo "brief: $brief"
28
- echo "base: $(git rev-parse HEAD)"
1
+ #!/usr/bin/env bash
2
+ # Begin one task of an inline plan execution in a single call: extract the
3
+ # task's brief (via subagent-driven-development's task-brief, so both skills
4
+ # share one workspace) and record BASE, the commit the task's review range is
5
+ # cut from. One tool call instead of two, because every call in an inline
6
+ # session is a turn that re-reads the whole context.
7
+ #
8
+ # Usage: task-start PLAN_FILE TASK_NUMBER
9
+ # Prints:
10
+ # brief: <path to the task's brief file>
11
+ # base: <full SHA of HEAD>
12
+ set -euo pipefail
13
+
14
+ if [ $# -ne 2 ]; then
15
+ echo "usage: task-start PLAN_FILE TASK_NUMBER" >&2
16
+ exit 2
17
+ fi
18
+
19
+ plan=$1
20
+ n=$2
21
+ sdd="$(cd "$(dirname "$0")/../../subagent-driven-development/scripts" && pwd)"
22
+
23
+ out=$("$sdd/task-brief" "$plan" "$n")
24
+ brief=$(printf '%s\n' "$out" | sed -n 's/^wrote \(.*\): [0-9][0-9]* lines$/\1/p')
25
+ [ -n "$brief" ] || { echo "task-brief did not report a path: $out" >&2; exit 1; }
26
+
27
+ echo "brief: $brief"
28
+ echo "base: $(git rev-parse HEAD)"
@@ -1,23 +1,11 @@
1
1
  ---
2
2
  name: gemini-collaborator
3
- description: Coordinate with Google Antigravity / Gemini CLI (agy.exe) to cross-validate changes, request second opinions on architecture/diffs, or execute adversarial code reviews.
3
+ description: Coordinate with Google Antigravity / Gemini CLI (agy.exe) via Herdr to cross-validate changes, request second opinions on architecture/diffs, or execute adversarial code reviews.
4
4
  ---
5
5
 
6
6
  # Gemini Collaborator Skill
7
7
 
8
- This skill enables Claude Code to orchestrate tasks with Google Antigravity / Gemini CLI (`agy.exe`), delegating complex architectural reasoning, diff reviews, codebase scans, and adversarial validation.
9
-
10
- ## Real-Time Visibility in Herdr (Right Split & Auto-Close)
11
-
12
- By default when running inside Herdr, all bridge invocations automatically:
13
- - **Split the calling screen to the right**: Creates a side-by-side split pane in your current Herdr window.
14
- - **Display a prominent request banner**:
15
- - Caller AI $\rightarrow$ Target Model (Claude Code $\rightarrow$ Antigravity / Gemini)
16
- - Action & Working Directory
17
- - **`>> WHAT IS BEING REQUESTED:`** A highlighted box showing the exact prompt/instructions.
18
- - Summary of modified files (`git diff --stat`) for code reviews.
19
- - **Stream Gemini's live execution**: Real-time display of reasoning, tool calls, and output right in that split pane.
20
- - **Auto-close when complete**: Automatically closes the right pane after finishing and collapses your screen back to normal, returning the captured output to Claude Code.
8
+ This skill enables Claude Code to orchestrate tasks with Google Antigravity / Gemini CLI (`agy.exe`) directly through native Herdr multiplexing, delegating complex architectural reasoning, diff reviews, codebase scans, and adversarial validation.
21
9
 
22
10
  ## When to Activate
23
11
 
@@ -25,59 +13,39 @@ By default when running inside Herdr, all bridge invocations automatically:
25
13
  - After making non-trivial code edits or architectural refactors, to verify there are no hidden regressions from Gemini's perspective.
26
14
  - When facing ambiguous design choices where Gemini's large context window and deep multi-perspective reasoning help pick the best approach.
27
15
 
28
- ## Available Actions via `tools/gemini-bridge.ps1`
16
+ ## Orchestration Method: Native Herdr Agent (`herdr` skill)
29
17
 
30
- Run commands using the project PowerShell helper script:
18
+ When running inside Herdr (`$env:HERDR_ENV = "1"`), communicate directly using Herdr's native agent orchestration. This maintains a persistent session across turns, preserves context, respects interactive approvals (`blocked`), and handles bracketed paste:
31
19
 
32
- ### 1. Adversarial Code Review of Git Diff
33
- Splits the screen to the right, runs Antigravity review, streams live, and auto-closes:
34
- ```powershell
35
- powershell -ExecutionPolicy Bypass -File tools/gemini-bridge.ps1 review
20
+ ### 1. Fast Path: Re-use Active Idle Agent
21
+ Before splitting a pane, check if a Gemini/Antigravity agent is already running and idle:
22
+ ```bash
23
+ herdr agent list
36
24
  ```
37
- To provide specific review instructions:
38
- ```powershell
39
- powershell -ExecutionPolicy Bypass -File tools/gemini-bridge.ps1 review "Focus on security vulnerabilities, memory leaks, and concurrency edge cases"
25
+ If an agent named `gemini-reviewer` or `agy` is idle or done, prompt it directly:
26
+ ```bash
27
+ herdr agent prompt gemini-reviewer "Review the latest changes." --wait --timeout 120000
40
28
  ```
41
29
 
42
- ### 2. Consult Gemini for Architectural Second Opinion
43
- Sends an inquiry or architectural prompt directly to Gemini in a right-split pane:
44
- ```powershell
45
- powershell -ExecutionPolicy Bypass -File tools/gemini-bridge.ps1 ask "How should we structure this module's caching and database connection pooling?"
30
+ ### 2. Spawning a New Gemini Agent (Single-Execution Batch)
31
+ If no active Gemini agent exists, batch pane creation and agent startup in a single command:
32
+ ```bash
33
+ PANE_ID=$(herdr pane split --current --direction right --cwd "$PWD" --no-focus | jq -r '.result.pane.pane_id')
34
+ herdr agent start gemini-reviewer --kind agy --pane "$PANE_ID"
35
+ herdr agent prompt gemini-reviewer "Review git diff for security regressions and type errors." --wait --timeout 120000
46
36
  ```
47
37
 
48
- ### 3. Scan Monorepo or Codebase
49
- Instructs Gemini to scan the repository for specific patterns or consistency with doctrine:
50
- ```powershell
51
- powershell -ExecutionPolicy Bypass -File tools/gemini-bridge.ps1 scan "Find all authentication token refresh and session handling references"
38
+ ### 3. Read Findings into Context
39
+ ```bash
40
+ herdr agent read gemini-reviewer --source recent-unwrapped --lines 120
52
41
  ```
53
42
 
54
- ### 4. Interactive Handoff to Gemini in a Right Split Pane
55
- Launches a full interactive Antigravity CLI session in a right split pane pre-loaded with a task:
56
- ```powershell
57
- powershell -ExecutionPolicy Bypass -File tools/gemini-bridge.ps1 interactive "Let's refactor the data access layer together"
43
+ ### 4. Interactive Collaboration
44
+ When an interactive session or clarification is needed, send keys or prompt interactively:
45
+ ```bash
46
+ herdr agent send-keys gemini-reviewer enter
58
47
  ```
59
48
 
60
- ### Display & Control Flags:
61
- - *(default)*: Splits the current Herdr screen to the right, streams live, and auto-closes when done.
62
- - `-KeepOpen`: Keeps the split pane open after completion instead of auto-closing.
63
- - `-Tab`: Opens in a new Herdr tab instead of splitting the screen.
64
- - `-Window`: Forces a standalone OS console window instead of Herdr.
65
- - `-Headless`: Runs silently in the background (invokes `agy -p` directly).
66
- - `-NoFocus`: Prevents switching focus to the right pane.
67
- - `-Timeout <seconds>`: Custom timeout in seconds (default: 300).
68
-
69
- ## Fallback Direct CLI Invocations (Headless / Standalone)
70
-
71
- If `tools/gemini-bridge.ps1` is not present in the workspace, Claude Code can invoke `agy` directly from the terminal:
72
- - **Headless prompt**:
73
- ```powershell
74
- agy --dangerously-skip-permissions -p "Review the latest changes in git diff and identify potential regressions."
75
- ```
76
- - **Interactive session**:
77
- ```powershell
78
- agy --dangerously-skip-permissions -i "Let's work through this problem together."
79
- ```
80
-
81
49
  ## Guidelines for Claude Code
82
50
 
83
51
  - **Synthesize, Don't Blindly Echo**: When Gemini returns feedback, analyze its critiques against the project's requirements, architectural documentation, and existing patterns.