benchflow-traj-upload

benchflow-traj-upload is a skill for Claude Code, Codex from benchflow-ai/benchflow. It costs 98 tokens per session (1,860 once invoked), scanned A, original, Apache-2.0.

A guided workflow for finding a local Claude Code or Codex session, opening its BenchFlow trajectory viewer, and submitting it after a person approves it. A trajectory is a recorded agent session that can be reviewed and shared for evaluation.

In plain words
What is it for?
Use it to set up BenchFlow, find and choose a recent session, review it in a local viewer, approve it, and report whether the submission was new or already recorded.
Why use it?
It keeps the human in control of what is submitted and avoids exposing private upload links or requiring them to run command-line instructions themselves.

Skill for Claude CodeCodex

Written for Claude Code and Codex: allowed-tools in frontmatter, but also reads ~/.codex or $CODEX_HOME. Also seen: reads .claude/ paths; mentions subagents; mentions Claude Code.

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/benchflow-ai/benchflow/benchflow-traj-upload
Any agent
npx skills add benchflow-ai/benchflow --skill benchflow-traj-upload
Clone the repo
git clone --depth 1 https://github.com/benchflow-ai/benchflow

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for benchflow-traj-upload

README.md
[![agentmods](https://agentmods.dev/badge/skills/benchflow-ai/benchflow/benchflow-traj-upload.svg)](https://agentmods.dev/skills/benchflow-ai/benchflow/benchflow-traj-upload)
Your own site
<a href="https://agentmods.dev/skills/benchflow-ai/benchflow/benchflow-traj-upload"><img src="https://agentmods.dev/badge/skills/benchflow-ai/benchflow/benchflow-traj-upload.svg" alt="Measured on agentmods" height="20"></a>
Per session 98 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,860 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 5.1 $0.00098 $0.01860
Opus 5 $0.00049 $0.00930
Sonnet 5 $0.00020 $0.00372
Haiku 4.5 $0.00010 $0.00186

Measured 6d ago against content hash e390d7cf12d8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

benchflow-traj-upload 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 6d 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.

.agents/skills/benchflow-traj-upload/SKILL.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Submit a trajectory

The human copied a line into this chat so you would do the work. They should not run BenchFlow commands. You find a local session, open the viewer, wait until they like it, then you upload.

Do not print broker URLs, Azure blob URLs, or a "run this yourself" command. Those leak private inbox paths and turn a paste-to-agent flow back into a CLI.

For operating, testing, or debugging the upload pipeline itself (dry runs, manifests, promotion checks), use benchflow-traj-upload-ops instead.

Workflow

1. setup     → ensure the latest benchflow is installed
2. discover  → list recent local Claude / Codex / trial sessions
3. pick      → user chooses one (or confirms your recommendation)
4. view      → open the viewer with --confirm and give them the localhost URL
5. confirm   → wait for the Approve button (or their chat reply)
6. submit    → you upload; report Submitted / Already submitted + digest
7. persist   → if this repo has no local copy of this skill, write one

Step 1 — Setup

Always make sure the latest BenchFlow is installed before anything else — bench traj setup and the session-JSONL viewer only exist in 0.7.1+:

uv tool install --python 3.12 --upgrade benchflow

If uv reports Executables already exist, rerun with --force. Verify that bench --version reports at least 0.7.1. If the installed CLI lacks bench traj setup or cannot open a session JSONL in the viewer, upgrade first rather than working around it.

Step 2 — Discover

Prefer the listing the CLI already knows how to make:

bench traj setup --list

If that command is missing, search these locations and skip nested subagents/ files unless the user asks:

  • Claude Code: ~/.claude/projects/**/*.jsonl
  • Codex: ~/.codex/sessions/**/*.jsonl and ~/.codex/archived_sessions/*.jsonl
  • Cursor: agent transcripts at ~/.cursor/projects/*/agent-transcripts/**/*.jsonl
  • OpenCode: newer versions keep sessions in a SQLite database at ~/.local/share/opencode/opencode.db (run opencode db path to confirm); older versions used JSON files under ~/.local/share/opencode/storage/session/
  • BenchFlow trials: jobs/**/trajectory/ or a directory with turn*.txt

Read the full file on GitHub · 182 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. 6d ago First seen · 182 lines · 98 tokens per session scan A e390d7cf12d8

Subscribe to this mod's changes

benchflow-traj-upload is a skill published in the GitHub repository benchflow-ai/benchflow (340 stars, last pushed yesterday), licensed Apache-2.0. It adds 98 tokens to every session and 1,860 once invoked, about $0.0005 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.

Related

Other skills, from other repositories

lastlight-evals

Scaffold, configure and run a Last Light EVALS workspace — the harness that runs Last Light's real workflows against a mocked GitHub and grades them deterministically. Use when the user wants to "set up / scaffold Last Light Evals", "create an evals workspace or instance", "run evals", "compare models", or author new…

nearform/lastlight · 185 tokens

desktop-principles

Desktop-specific UX principles - hover states, pointer precision, keyboard shortcuts, multi-window, focus management. Covers macOS, Windows, Linux, web desktop.

Jwuthri/Tracely-ai · 36 tokens

canvas-generative

Algorithmic and generative art with Canvas 2D - particles, flow fields, noise, fractals, L-systems.

Jwuthri/Tracely-ai · 30 tokens

ceo-setup

One-time onboarding for the executive/manager commitment workflow — delegation-heavy, meeting prep, decision capture, morning and evening digests. Creates a commitments project and installs two dashboard widgets. After successful setup this skill is excluded from selection until the marker file is deleted.

suyoumo/ClawProBench · 60 tokens

portfolio

Cross-chain DeFi portfolio discovery, rebalancing suggestions, and NEAR Intent construction. Activates when the user pastes a wallet address or asks about yield/positions/rebalancing. Bootstraps a per-user "portfolio" project, aggregates positions across all the user's addresses inside one project, and offers a…

suyoumo/ClawProBench · 69 tokens

content-creator-setup

One-time onboarding for the content creator workflow — content pipeline stages, trend expiration, cross-platform cascades, heavy idea parking. After successful setup this skill is excluded from selection until the marker file is deleted.

suyoumo/ClawProBench · 47 tokens