Skip to content
cccskills
FreePublished on GitHub

call-glm

Hand a self-contained task from a Claude session to a GLM agent running in Zed through agent-mail. Use it to request a second opinion or bulk pass without accessing the z.ai API key, then verify the peer agent’s reply before acting on it.

OriginalShow the original description

How a Claude session hands work to a GLM agent running in Zed, through agent-mail, instead of touching the z.ai API key. Use when you want a GLM second opinion, a cheap bulk pass, or to offload a bounded task to a Zed native thread, and whenever you are tempted to call api.z.ai yourself.

katopz/katgpt-rs141 stars(GitHub 141 + cccskills 0)

Run score Not tested

See how to install

Good for

  • Ask a GLM agent for a second opinion
  • Delegate a bounded task from a Claude session
  • Request a bulk review through a Zed native thread

Files included (1)

  • SKILL.md2.9 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

boundary-guard

FreeSummary

Audit how responsibilities and dependencies are divided across a multi-repository game workspace. The skill checks each repository’s BOUNDARY.md contract, dependency relationships, and code patterns to identify misplaced game logic, view-layer behavior, FFI data, or outdated boundary records.

  • Checking boundaries before adding game logic
  • Reviewing a change to a game system or view layer
  • Finding dependency or ownership drift
katopz/katgpt-rs141Updated Nov 19, 2026

build-hygiene

FreeSummary

Use this skill to review Rust build profiles, binary size, and the edit-to-result loop. It sets measurement rules for comparing build approaches, keeping target directories manageable, and reporting performance claims with their profile and test conditions.

  • Reviewing Cargo profiles after adding dependencies
  • Investigating slow builds or a growing target directory
  • Comparing development-loop options with measured results
katopz/katgpt-rs141Updated Nov 17, 2026

calibration-gate

FreeSummary

Plan and review calibration or tuning runs using evidence counted for each group and decision rules set before testing. Keep calibration separate from production checks, attach machine and workload details to measurements, and report calibrated, fallback, sparse, or undecided results honestly.

  • Prepare a tuning run with decision rules set in advance
  • Review why a calibration run failed before testing
  • Classify results by group and record the evidence behind each metric
katopz/katgpt-rs141Updated Nov 16, 2026

disk-clean

FreeSummary

Inspect disk usage across a development workspace, caches, model folders, and removable storage. Identify regenerable caches and report valuable or irreplaceable files separately, with before-and-after measurements when cleanup is performed.

  • Finding what is using disk space
  • Freeing space before a large build or training run
  • Reviewing cleanup candidates and their importance
katopz/katgpt-rs141Updated Nov 19, 2026

doc-sync

FreeSummary

Use this skill to bring repository documentation up to date with work that has landed in Git. It compares commits with the last documentation update, checks plans and issues, and guides updates to the README, docs, and boundary records.

  • Updating docs after a plan or issue is completed
  • Checking whether README claims match shipped work
  • Reviewing documentation coverage during a quarterly cleanup
katopz/katgpt-rs141Updated Nov 19, 2026

Check claims about feature flags against production code, build configuration, downstream consumers, and related documentation. The skill helps distinguish a genuine wiring problem from an intentional layer split, find stale status comments across multiple surfaces, and settle whether a gate applies through the full build chain.

  • Verifying whether a feature flag reaches production code
  • Checking for stale comments after a flag promotion
  • Distinguishing an intentional layer split from a discrepancy
katopz/katgpt-rs141Updated Nov 15, 2026

See all skills by katopz

Report a problem with this skill