You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Round-up of 8 unresolved threads on PR #346.
Test stability (Copilot ×2)
- __tests__/cli-subcommands.test.js: derive expected count from
loadMarketplace().plugins.length and assert anchor plugins are
present. Old hardcoded '20 plugin(s) found' / 'with 20 plugins' was
destined to break on every marketplace addition.
Generator parity (Copilot, Gemini HIGH)
- scripts/generate-docs.js: generateAgentCounts now uses static
fallbacks when local discovery is empty (post-graduation reality).
Auto-gen block in docs/reference/AGENTS.md now reports
'49 agents across 20 plugins (18 have agents)' instead of the
meaningless '10 agents across 0 plugins'.
- __tests__/generate-docs.test.js: tests track the same fallback.
Count accuracy across docs (Copilot, Gemini MEDIUM)
- AGENTS.md, docs/reference/AGENTS.md: prose now says
'49 agents across 20 plugins (18 have agents)'. Earlier '47 / 17'
prose was a pre-existing inaccuracy that this PR quietly corrects -
ship has 1 agent (release-agent.md) so it belongs in the
'have agents' bucket; only gate-and-ship and zig-lsp don't.
- docs/reference/AGENTS.md L36: file-based agent count 37 -> 39
(matches sum of STATIC_PLUGIN_AGENT_COUNTS).
- site/index.html L752: 'Skills across 20 Plugins' -> '19 Plugins'.
zig-lsp contributes 0 skills, so the count of skill-contributing
plugins is unchanged. Reverting Gemini's flagged inaccuracy.
- docs/ARCHITECTURE.md L277: sync-pipeline language softened to
'19 plugin repos today; zig-lsp sync wire-up pending', matching
the open checkbox in docs/ORG_ARCHITECTURE.md.
- site/ux-spec.md L231: pre-existing typo Skills=40 -> 41.
PR description has stale 'deliberately not updated: generate-docs.js'
- will update separately as a PR comment once these land.
Copy file name to clipboardExpand all lines: docs/ARCHITECTURE.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -275,7 +275,7 @@ Research documents informing the implementation (in `agent-docs/`):
275
275
276
276
**Update workflow:**
277
277
1. Edit files in `lib/` (canonical source in this repo)
278
-
2. Push to main — agent-core sync pipeline automatically opens PRs in all 20 plugin repos
278
+
2. Push to main — agent-core sync pipeline automatically opens PRs in 19 plugin repos today (`zig-lsp` sync wire-up still pending; tracked in docs/ORG_ARCHITECTURE.md)
279
279
3. Merge those PRs in each plugin repo
280
280
4. Publish agentsys: `npm version patch && npm publish`
Copy file name to clipboardExpand all lines: docs/reference/AGENTS.md
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -3,7 +3,7 @@
3
3
Complete reference for all agents in AgentSys.
4
4
5
5
<!-- GEN:START:agents-counts -->
6
-
**TL;DR:**10 agents across 0 plugins (1 have agents). opus for reasoning, sonnet for patterns, haiku for execution. Each agent does one thing well. <!-- AGENT_COUNT_TOTAL: 10-->
6
+
**TL;DR:**49 agents across 20 plugins (18 have agents). opus for reasoning, sonnet for patterns, haiku for execution. Each agent does one thing well. <!-- AGENT_COUNT_TOTAL: 49-->
7
7
<!-- GEN:END:agents-counts -->
8
8
9
9
---
@@ -24,7 +24,7 @@ Complete reference for all agents in AgentSys.
24
24
25
25
## Overview
26
26
27
-
AgentSys uses 47 specialized agents across 20 plugins (17 have agents; ship and gate-and-ship are commands-only; zig-lsp is a config-only LSP plugin with no commands or agents). Each agent is optimized for a specific task and assigned a model based on complexity:
27
+
AgentSys uses 49 specialized agents across 20 plugins (18 have agents; gate-and-ship is commands-only; zig-lsp is a config-only LSP plugin with no commands or agents). Each agent is optimized for a specific task and assigned a model based on complexity:
28
28
29
29
| Model | Use Case | Cost |
30
30
|-------|----------|------|
@@ -33,7 +33,7 @@ AgentSys uses 47 specialized agents across 20 plugins (17 have agents; ship and
33
33
| haiku | Mechanical execution, no judgment | Low |
34
34
35
35
**Agent types:**
36
-
-**File-based agents** (37) - Defined in `plugins/*/agents/*.md` with frontmatter <!-- AGENT_COUNT_FILE_BASED: 37-->
36
+
-**File-based agents** (39) - Defined in `plugins/*/agents/*.md` with frontmatter <!-- AGENT_COUNT_FILE_BASED: 39-->
37
37
-**Role-based agents** (10) - Defined inline via Task tool with specialized prompts <!-- AGENT_COUNT_ROLE_BASED: 10 -->
return`**TL;DR:** ${totalAgents} agents across ${plugins.length} plugins (${pluginCount} have agents). opus for reasoning, sonnet for patterns, haiku for execution. Each agent does one thing well. <!-- AGENT_COUNT_TOTAL: ${totalAgents} -->`;
400
+
return`**TL;DR:** ${totalAgents} agents across ${totalPlugins} plugins (${pluginsWithAgentsCount} have agents). opus for reasoning, sonnet for patterns, haiku for execution. Each agent does one thing well. <!-- AGENT_COUNT_TOTAL: ${totalAgents} -->`;
0 commit comments