Gajae Code is an external coding-agent harness that works inside repositories or worktrees and guides an agent through planning, review, approval, and code changes. It is for developers who want to use an existing coding-plan subscription while controlling agent work from a terminal, phone, or bot. The catalogue entries provide extensions such as skills, agents, commands, plugins, and MCP integrations for the harness.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Yeachan-Heo/gajae-code --skill gjc-sdk-operategit clone --depth 1 https://github.com/Yeachan-Heo/gajae-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/yeachan-heo/gajae-code/gjc-sdk-operate)<a href="https://agentmods.dev/skills/yeachan-heo/gajae-code/gjc-sdk-operate"><img src="https://agentmods.dev/badge/skills/yeachan-heo/gajae-code/gjc-sdk-operate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yeachan-heo/gajae-code/gjc-sdk-operate"><img src="https://agentmods.dev/badge/skills/yeachan-heo/gajae-code/gjc-sdk-operate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.00740 |
| Opus 5 | $0.00013 | $0.00370 |
| Sonnet 5 | $0.00005 | $0.00148 |
| Haiku 4.5 | $0.00003 | $0.00074 |
Grade A, and why
gjc-sdk-operate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GJC SDK approved operations
This skill is for trusted local scripts. Its approval challenge is a procedural safety policy, not a security boundary; SDK core retains lifecycle and attachment authority.
Before every operation
- Select an exact session ID through
gjc sdk session listor a caller-provided stable ID, then fail closed when the Broker cannot prove it available. - Use only
gjc sdk session raw query|control|global; never scan state roots, read endpoint credentials, or open raw per-session WebSockets. - Validate the operation against the allowlist below. Do not expose arbitrary operation passthrough.
- Pass all command data as argv values, never through a shell command string.
- For every lifecycle operation, show the exact operation and session target to the human through the external host.
- Obtain one explicit approval immediately before the call. Approval is single-use and becomes invalid if the operation, input, or target changes.
The templates emit a nonce-bearing, input-bound
APPROVE <session> <operation> <digest> <nonce>challenge and read the exact response once from the active process's standard input. Present it verbatim through the external host only after the human accepts that exact action. - On denial, cancellation, unavailable target, or changed input, send no CLI request.
- Render only bounded, redacted CLI JSON; discard raw CLI stderr.
Allowed per-session controls
turn.promptturn.steerturn.follow_upask.answerworkflow.gate_answertodo.replacesession.switchsession.rename
For workflow.gate_answer, use the durable workflow gate ID and pass expectedSessionId. Never use transient action_needed.id as durable authority.
Long-running prompts
The SDK prompt deadline is progress-aware: sdk.promptDeadlineMs (30 min, 60_000–86_400_000) is an inactivity lease renewed only by attributable tool_execution_start / tool_execution_end for the exact accepted commandId/turnId, bounded by sdk.promptMaxRuntimeMs (6 h default, 60_000–86_400_000, caps at 24 h). Persist session_id / turn_id from turn.prompt acceptance and reconcile with turn.result (Q26) rather than replaying blindly. Distinguish the bounded await_turn poll timeout_ms from the SDK terminal deadline; heartbeats, streaming text/thinking deltas, retries, and other-turn activity do not renew the lease.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 62 lines · 26 tokens per session scan A 084f63931547
gjc-sdk-operate is a skill published in the GitHub repository Yeachan-Heo/gajae-code (2,752 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 740 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…