A criterion may ask whether the spec builds

[gHashTag]

What is wrong

CRITERION_T27C_SUBCOMMANDS held five names, so every acceptance criterion
filed against gHashTag/t27 was a grep. A grep is satisfied by a file that never
builds: 388 of the 837 specs the oracle can reach generate Zig that does not
compile, and each of them satisfies every criterion ever written against it.

What changes

• test-report is admitted. It generates the reviewed spec and has Zig build and run that spec's own tests -- the oracle's question, asked before the oracle. The bee already has a shell in this container, so nothing here is newly reachable to it; what the allowlist prevents is a criterion rigging its own measurement by running a script from the commit under review, and test-report runs the compiler on one named spec.
• The readers beside it (coverage, lint, symbols, outline, exports, metrics, deadcode, orphans, spellcheck, validate-vacuity, xref, api-diff, diff, eval, ...) were measured on the corpus on 2026-09-20: everything except test-report (35 s) answers in under 30 s, most in under one.
• dupes (~50 s) and check-calls (~140 s) are refused, with tests pinning the refusal. Past the 60 s per-command timeout a criterion is inconclusive, which spends a review and establishes nothing. A bee's own shell has no allowlist and can run either.

Why it is safe

Everything admitted reads a spec and prints. Nothing on the list writes to the
tree, opens a socket, drives hardware, or runs a script from the commit under
review -- the shape that kept battery, gates, silicon, fmt, rename
and seal off it, and that the tests continue to pin.

Measured

$ t27c lint specs/git/diff.t27 | head -1
WARN: fn 'diff' has no test or invariant
$ t27c test-report specs/git/diff.t27 2>&1 | grep -c BLOCKED
1

The companion change in gHashTag/t27 (#4301) puts the same commands in front of
the bees that write the specs.

🤖 Generated with Claude Code

[github-actions]

❌ Tests failed — 11/2431 failed

Suite	Passed	Failed	Skipped
✅ agent	87/87	0	0
✅ build	9/9	0	0
✅ cdp-protocol	5/5	0	0
✅ eval	93/93	0	0
✅ server-agent	272/272	0	0
❌ server-api	1248/1311	6	57
✅ server-browser	6/6	0	0
✅ server-integration	10/11	0	1
✅ server-lib	279/279	0	0
❌ server-pglive	1/3	2	0
✅ server-root	68/68	0	0
✅ server-skills	31/31	0	0
❌ server-tools	239/242	3	0
✅ shared	14/14	0	0

Failed tests

• server-api — the note says who committed the branch, where it can be read > puts the salvage fact where the 1500-character cap cannot cut it
• server-api — a turn the provider killed does not spend the issue > is judged and sent back, but charges neither counter
• server-api — deriving work the repository already measured > carries the command that produced it
• server-api — deriving work the repository already measured > gives every candidate a path that can be a boundary
• server-api — deriving work the repository already measured > proposes nothing from code this project does not own
• server-api — salvaging a turn that ended with its work uncommitted > refuses a worktree holding an unmerged path
• server-pglive — the migration block, applied to a real PostgreSQL > creates every object it promises, and survives a second boot
• server-pglive — the live gate > ran, or its absence is on the record
• server-tools — navigation tools > new_hidden_page opens a hidden tab
• server-tools — navigation tools > show_page restores a hidden page to visible
• server-tools — window tools > create_hidden_window creates and closes a hidden window

View workflow run