context-bridge

A local memory tool for Claude Code that keeps task progress between work sessions and notices when work is repeating or stuck.

In plain words
What is it for?
Save goals, progress, blockers, and next steps for later sessions. Review plans and retry earlier approaches when a task keeps reappearing.
Why use it?
It prevents you from having to reconstruct context after a session ends and highlights stalled work or repeated attempts.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/pushkalkumar/context-bridge/skill
Any agent
npx skills add pushkalkumar/context-bridge --skill skill
Clone the repo
git clone --depth 1 https://github.com/pushkalkumar/context-bridge

Made for: Claude Code, Codex.

Per session 89 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,041 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00089 $0.01041
Opus 5 $0.00044 $0.00521
Sonnet 5 $0.00018 $0.00208
Haiku 4.5 $0.00009 $0.00104

Measured yesterday against content hash baa2f75cb46e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

context-bridge 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.

skill/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

context-bridge

Persistent task memory across Claude Code sessions: checkpoints, stagnation detection, attempt replay, and planning. The backend auto-starts when a session begins and hooks checkpoint automatically. Everything you do goes through two CLI commands — never hand-build HTTP calls (raw API, only if the CLI is unavailable: references/api.md).

Record progress: context-bridge sync

Run at these moments — not every message:

  • The user states or changes the session goal.
  • A task completes and the next step is known.
  • A blocker is hit (pass the exact error via --blocker).
context-bridge sync \
  --goal "Ship JWT auth" \
  --task "Implement /login endpoint" \
  --progress "/register done; /login returns 500 on bcrypt verify" \
  --blocker "error: cannot find module 'bcrypt'" \
  --next "Fix bcrypt import, then sign HS256 token"

--blocker is repeatable. Project ID and git state are gathered automatically.

The output is the plan. Act on it:

  • Plan: — execute it; on conflict with the user request, see Conflicts.
  • Priority: — the active constraint for the session.
  • stagnation N with N ≥ 3 — apply the stagnation protocol (below).
  • confidence below 75% — surface the uncertainty before proceeding.
  • ⚠ Blocker class: — address that blocker type before retrying.
  • ⚠ Decompose: — split into subtasks of 30 minutes or less first.
  • rule-based source — follow the plan exactly, do not improvise.

Record events: context-bridge event

context-bridge event failure --attempted "session cookies" --because "stateless API requirement"
context-bridge event adr --decision "HS256 signing" --reason "single service" --tradeoff "rotation needs redeploy"
context-bridge event outcome --result "auth suite green" --impact "auth feature complete"

Record failure when abandoning an approach (feeds blocker matching), adr on architecture decisions, outcome when a significant result lands. Goal and task default to the latest checkpoint's — override with --goal/--task if the event belongs elsewhere.

Read the full file on GitHub · 102 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. yesterday First seen · 102 lines · 89 tokens per session scan A baa2f75cb46e

Subscribe to this mod's changes

context-bridge is a skill published in the GitHub repository pushkalkumar/context-bridge (2 stars, last pushed 1mo ago), licensed MIT. It adds 89 tokens to every session and 1,041 once invoked, about $0.0004 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.