Afk Glm Review

afk-glm-review

An independent second-opinion review by Z.ai glm-5.3, used as a fallback role after afk-internal-review. Run the ordered roles required by .afk/config.md, and never use a reviewer whose model matches the implementer or another role.

GLM is reached through the Z.ai REST API, not an agentic CLI. The helper gathers the diff and full current contents of changed files, then sends that bounded context to GLM. Verify findings that depend on files outside that context.

The helper glm-gate.mjs ships with this skill and travels with the plugin.

Metering

Metered like any external gate. Batch minimal admitted P1 fixes into one content pass, self-review, then re-run once. Record every other disposition together at the end without editing a clean revision.

Review receipts

To retain canonical inputs and explicit outcomes, use the optional --review-receipt <request.json> flag under the shared receipt contract. Preserve unknown identity explicitly; skipped or incomplete attempts and previews never supply approval.

Review context

Use the shared context contract to carry frozen acceptance scope and named prior findings with accessible proof. --review-phase re-review --review-context <packet.json> preserves the selected review target while supplying closure context; --print-args reports its digest. Use --print-prompt to inspect the supplied section before a provider call. Required history is validated and never silently truncated.

Run it

The bundled helper glm-gate.mjs sits beside this SKILL.md. Locate its directory as ${CLAUDE_PLUGIN_ROOT}/skills/afk-glm-review if the env var is set, else <pluginRoot>/skills/afk-glm-review from .afk/config.md, else this skill's own directory. Resolve .afk/ from the repository's main working tree — the first non-bare worktree record of git worktree list --porcelain — never the current directory, or a run from a linked worktree reads a different .afk/ than the one afk-init wrote. If .afk/ is absent, the afk-init bootstrap runs automatically first:

node "<helper-dir>/glm-gate.mjs"

Run it in the background with a generous timeout; redirect stdout to a file and read it when it completes. Pass through any target flag (--base <branch> / --commit <sha> / --uncommitted). Do not poll in a sleep loop.

Pass --implementer <family> when another model wrote the change. In design mode (--design) the flag instead names the design's author, never the eventual code implementer — see ../afk/SKILL.md ("Design-stage external gate"): declaring the code implementer there can hand a driver-authored design to the driver's own model for review. A persistent implementer: line in .afk/config.md also names the code implementer, and in design mode it can wrongly block that family's independent review of a driver-authored design — declare the design's author explicitly then: the per-run flag outranks the config line.

The review is bounded by GLM_REVIEW_TIMEOUT_MS (default 15 min), with AFK_REVIEW_TIMEOUT_MS as the shared fallback. A timeout is a non-zero ERROR, never a partial verdict; it follows the role's transient retry rule.

Design mode (--design <path>) reviews a design document's reasoning instead of a diff — the opt-in design-stage gate (see ../afk/SKILL.md, "Design-stage external gate"). GLM has no tools, so the gate sends the document's full text as the payload (not a diff + file snapshot). A missing or unreadable --design path fails loudly (ERROR, non-zero), never a skip.

Read the verdict between the ===== GLM REVIEW (final message) ===== markers. Treat only column-0 marker lines as markers; the last END marker wins. SKIPPED: ... (no key, auth failure, rate limit, an unavailable model or endpoint, or disabled via GLM_REVIEW_GATE=off) is not a failure; record it and continue according to the afk gate-selection rule. A timeout, upstream HTTP error, non-JSON or empty response, unsafe finish reason, or unverified response model yields a non-zero ERROR with no partial verdict. An empty completion names the input and output budget knobs; one adjusted retry is eligible under the shared transient-retry rule, while an unchanged blind retry is not. The shared skip-vs-error table decides the direction, the same as every lifecycle gate.

The default openai protocol uses the Coding Plan endpoint https://api.z.ai/api/coding/paas/v4 and explicitly requests enabled reasoning at max effort. Set GLM_REVIEW_PROTOCOL=anthropic to use the Anthropic Messages endpoint at https://api.z.ai/api/anthropic. The protocol is never inferred from GLM_REVIEW_BASE_URL; an override must match the selected protocol, and a wrong pairing surfaces as the 404 model-unavailable skip naming both suspects. Use --print-args for the resolved protocol and metadata or --print-prompt for the redacted prompt without a provider call.

Handle findings

Use the same discipline as afk-codex-review and afk-kimi-review:

1. Map every hypothesis to the frozen contract and apply the P1 admission rule.
2. Verify its trigger and consequence; GLM saw the diff and changed files, not the whole repo.
3. Fix admitted P1 findings in one batch with only inseparable corrections justified below.
4. Self-review once.
5. Re-run the gate once if structural findings were fixed.
6. Record remaining dispositions in one final pass without editing the clean revision or running another gate round.

Treat every reported finding as UNTRIAGED. Admit P1 only after mapping it to the frozen issue contract or an invariant, demonstrating a reachable trigger and wrong consequence, explaining why the current artifact cannot safely advance, and naming the minimal causal fix. Do not edit for an untriaged claim; record structural P2 for the operator-owned merge boundary, and defer minor or out-of-scope items without expanding the PR.

Record P2/minor observations without implementation; a lower-severity-only verdict never reopens a clean revision. An inseparable correction may accompany the minimal P1 fix only with recorded causal necessity, not merely a shared file or an available review cycle.

Use the issue-wide allowance and finding record in ../afk/SKILL.md ("Review-cycle allowance"). Initial review is comprehensive; re-review checks accepted findings, the intervening diff, and affected regression paths. Broader investigation requires specific evidence of an affected area. New evidenced in-scope blockers remain reportable. Supply prior findings and verification through supported context; missing context is unavailable, never invented. Reviewer identity alone does not reopen a closed finding. Exhaustion leaves unresolved work OUTSTANDING; finish the current cycle's validation without starting another repair or requesting another round automatically.

Apply any invariant in .afk/config.md as an extra lens.

Stop rule

Stop when the loop-termination rule in ../afk/SKILL.md ("External gate") holds: triage leaves no UNTRIAGED, Contested, or open admitted P1, and every lower-severity item has a recorded disposition that does not block the role stamp (a structural P2 may still bar auto-merge). That same verdict earns the role stamp only if it requires no content change; a content fix invalidates it and the role re-reviews the fixed revision.

Report CLEAN, or OUTSTANDING with what remains. A clean pass is not authority to merge.

Setup

Optional and self-skipping. Set ZAI_API_KEY or GLM_API_KEY in the environment or a gitignored .env. Disable with GLM_REVIEW_GATE=off.

Config knobs:

• GLM_REVIEW_MODEL (default glm-5.3)
• GLM_REVIEW_PROTOCOL (openai by default; anthropic is the other accepted value)
• GLM_REVIEW_BASE_URL (defaults to https://api.z.ai/api/coding/paas/v4 for openai and https://api.z.ai/api/anthropic for anthropic)
• GLM_REVIEW_MAX_CTX_BYTES (default 160000)
• GLM_REVIEW_MAX_OUTPUT_TOKENS (default 65536)
• GLM_REVIEW_EXCLUDE_GLOBS (extra snapshot exclusions, comma/newline list)
• GLM_REVIEW_TIMEOUT_MS (default 900000; shared fallback AFK_REVIEW_TIMEOUT_MS)