everyinc--compound-engineering-plugin
wehub-resource-sync
77 行
3.6 KiB
TypeScript
77 行
3.6 KiB
TypeScript
import { readFileSync } from "fs"
|
|
import path from "path"
|
|
import { describe, expect, test } from "bun:test"
|
|
|
|
const AGENT_PATH = path.join(
|
|
process.cwd(),
|
|
"skills/ce-compound/references/agents/session-historian.md",
|
|
)
|
|
const COMPOUND_SKILL_PATH = path.join(
|
|
process.cwd(),
|
|
"skills/ce-compound/SKILL.md",
|
|
)
|
|
const AGENT_BODY = readFileSync(AGENT_PATH, "utf8")
|
|
const COMPOUND_SKILL_BODY = readFileSync(COMPOUND_SKILL_PATH, "utf8")
|
|
|
|
// Regression guard for https://github.com/EveryInc/compound-engineering-plugin/issues/794.
|
|
//
|
|
// `session-historian.md` runs in subagent context (dispatched by `ce-compound`
|
|
// Phase 1). Claude Code does not permit
|
|
// subagents to invoke the `Skill` tool — the call hangs at "Initializing…"
|
|
// indefinitely, eventually surfacing to the orchestrator as a spurious
|
|
// "user doesn't want to proceed with this tool use" rejection
|
|
// (anthropics/claude-code#38719).
|
|
//
|
|
// The fix keeps all script orchestration in ce-compound's main context,
|
|
// keeping this prompt synthesis-only so it reads
|
|
// pre-extracted scratch files via the platform's native file-read tool.
|
|
//
|
|
// This test locks the no-Skill-from-subagent invariant: the agent's body
|
|
// must not instruct any `Skill(...)` invocation. Silent regression here
|
|
// reintroduces the deadlock.
|
|
describe("session-historian prompt no-Skill-tool regression guard", () => {
|
|
test("agent body does not instruct Skill(ce-session-inventory) calls", () => {
|
|
expect(AGENT_BODY).not.toMatch(/Skill\(\s*["'`]?ce-session-inventory/)
|
|
})
|
|
|
|
test("agent body does not instruct Skill(ce-session-extract) calls", () => {
|
|
expect(AGENT_BODY).not.toMatch(/Skill\(\s*["'`]?ce-session-extract/)
|
|
})
|
|
|
|
test("agent body does not contain the broken-pattern prose fingerprint", () => {
|
|
expect(AGENT_BODY).not.toMatch(/Invoke them through the Skill tool/i)
|
|
})
|
|
|
|
test("agent body does not instruct any Skill(...) tool-call expression", () => {
|
|
// Belt-and-suspenders: any literal `Skill(...)` tool-call form in the
|
|
// agent body would deadlock under the same constraint. The agent's
|
|
// contract is "read paths via native file-read; never invoke Skill."
|
|
// Backtick-quoted prose mentions like `Skill` are fine — only literal
|
|
// call expressions are flagged. Match `Skill(` followed by a non-space
|
|
// character (excluding the closing backtick that would mark a code span).
|
|
const skillCallPattern = /(?<!`)Skill\([^)`]/
|
|
const match = AGENT_BODY.match(skillCallPattern)
|
|
expect(
|
|
match,
|
|
`Agent body contains a literal Skill(...) tool-call expression: ${match?.[0]}. ` +
|
|
`Subagents cannot invoke the Skill tool in Claude Code (issue #794). ` +
|
|
`Use the platform's native file-read tool on pre-extracted paths instead.`,
|
|
).toBeNull()
|
|
})
|
|
|
|
test("ce-compound still dispatches the skill-local session historian prompt", () => {
|
|
expect(COMPOUND_SKILL_BODY).toContain("references/agents/session-historian.md")
|
|
expect(COMPOUND_SKILL_BODY).toContain("Do not dispatch a standalone agent by type/name")
|
|
})
|
|
|
|
test("ce-compound resolves repo filter before running session discovery", () => {
|
|
expect(COMPOUND_SKILL_BODY).toContain("REPO_ROOT=$(git rev-parse --show-toplevel")
|
|
expect(COMPOUND_SKILL_BODY).toContain('REPO_NAME=$(basename "$REPO_ROOT")')
|
|
expect(COMPOUND_SKILL_BODY).toContain('discover-sessions.sh" "$REPO_NAME" "$SCAN_DAYS"')
|
|
expect(COMPOUND_SKILL_BODY).toContain('--cwd-filter "$REPO_ROOT"')
|
|
expect(COMPOUND_SKILL_BODY).not.toContain('discover-sessions.sh" <repo> <days>')
|
|
expect(COMPOUND_SKILL_BODY).not.toContain("--cwd-filter <repo>")
|
|
expect(COMPOUND_SKILL_BODY).not.toContain("<resolved repo root")
|
|
})
|
|
})
|