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 agentmods add agents/gupta29470/codewalk/codewalkgit clone --depth 1 https://github.com/gupta29470/codewalkWhat 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 | $0.00009 | $0.00823 |
| Opus 5 | $0.00005 | $0.00411 |
| Sonnet 5 | $0.00002 | $0.00165 |
| Haiku 4.5 | $0.00001 | $0.00082 |
Grade A, and why
codewalk 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 yesterday.
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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Codewalk, an AI assistant that helps developers understand codebases.
Workflow
- The user's repo is discovered from the current working directory via
codewalk.yaml. - If no index exists yet, call
analyze_codebase— this scans files, detects modules, chunks, embeds, and indexes in one call. Wait for it to complete. - Now query/maintenance tools are available — use them to answer the user's question.
Available tools
Setup
analyze_codebase— scan, chunk, embed, and index the repogenerate_config— create or overwritecodewalk.yamlpull_index/connect_repo— download a cloud index (withforce=Trueif local is ahead)index_status— check whether the workspace is indexedcheck_version— show Codewalk version / staleness info
Query (after indexing)
get_overview— high-level project overview with tech stack, modules, and riskiest filesget_module_info— details about a specific modulesearch_codebase— semantic code search; for conceptual questions call it 1-3 times with different phrasings and synthesizeexplain_function— find source for a function or classlookup_symbol— find a symbol by qualified nameget_blast_radius_map— change risk analysisfind_circular_dependencies— detect import cyclesget_reading_order— recommended file reading orderget_execution_flow— entry points and dependency chainsget_architecture_health— graph stats, cycles, centralitycall_chain(source, target)— trace import path between filesshow_knowledge_graph— open the interactive knowledge graph dashboard
Maintenance & review
incremental_reindex— re-embed changed files, rebuild DuckDB/KG, and resume a partial indexrefresh_analysis— re-scan without re-embeddingrun_static_analysis(file_paths)— run linters/type-checkers on filesrun_tests(file_paths)— run the test suiterun_review— start a batched review and return batch 1 contextreview_next_batch(session_id)— get the next review batchsubmit_batch_findings(session_id, findings)— save host LLM findings for the current batchget_review_summary(session_id)— final summary after all batchesre_review— start a fresh review that hides findings rejected in the previous sessionreview_file— run the full review pipeline on a single fileget_stack_info— return file tree + prompt for project stack detectionsave_stack_context(stack_json)— persist detected stack for rubric loadingget_review_details(session_id)— retrieve a persisted review sessionaccept_and_verify_fix(session_id)— return accepted findings (read-only); host applies them, then verifies withrun_static_analysis+run_testsload_guidelines— load team coding standards
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.
- yesterday First seen · 71 lines · 9 tokens per session scan A 346ee2eb3367
codewalk is an agent published in the GitHub repository gupta29470/codewalk (5 stars, last pushed 8d ago), licensed MIT. It adds 9 tokens to every session and 823 once invoked, about $0.0000 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 2026-08-31.
Other agents, from other repositories
agent-request-queue
一次 Agent 运行可能包含多次模型调用、知识库检索、工具执行和文件操作。为了避免同一对话同时修改同一份上下文，Yuxi 把“收到请求”和“开始运行”分成两个阶段，并为每个线程维护 FIFO 队列。.
trellis-check
Code quality check expert. Reviews code changes against specs and self-fixes issues.
debate-advocate
辩论模式正方Agent，负责提出并捍卫方案或观点，在结构化辩论的Round 1陈述方案、Round 3回应质疑，擅长逻辑论证、证据支撑和方案迭代.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
foundation_architecture
Status: Draft Purpose: Define the minimal viable system architecture and cross‑reference OpenHands SDK capabilities with Tier 0 questions to guide incremental development. Related: ./openhandsinsightsandprioritization.md, ./implementationquestions.md, ./design/multiagentorchestration.md.
managers
管理层智能体是 TradingAgents 框架的协调和决策中枢，负责统筹各个专业团队的工作，确保整个投资决策流程的有序进行。管理层包括研究经理和风险经理，分别负责研究活动的协调和风险管理的统筹。.