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 skills/nyktora/noctrace/jsonl-parsernpx skills add nyktora/noctrace --skill jsonl-parsergit clone --depth 1 https://github.com/nyktora/noctraceWrote 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/nyktora/noctrace/jsonl-parser)<a href="https://agentmods.dev/skills/nyktora/noctrace/jsonl-parser"><img src="https://agentmods.dev/badge/skills/nyktora/noctrace/jsonl-parser.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00032 | $0.01102 |
| Opus 5 | $0.00016 | $0.00551 |
| Sonnet 5 | $0.00006 | $0.00220 |
| Haiku 4.5 | $0.00003 | $0.00110 |
Grade A, and why
jsonl-parser 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 4d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code JSONL Parser Specification
File Location
Session logs are at ~/.claude/projects/{encoded-path}/*.jsonl where path encoding replaces / with - (e.g., /Users/jane/project becomes -Users-jane-project).
Additionally, ~/.claude/projects/{encoded-path}/sessions-index.json contains session metadata (summaries, message counts, timestamps).
Record Format
Each line is a JSON object with these common fields:
interface BaseRecord {
type: "user" | "assistant" | "system" | "result";
sessionId: string;
timestamp: string; // ISO-8601
uuid: string;
parentUuid: string | null;
version?: string;
cwd?: string;
gitBranch?: string;
}
Assistant Records
Contain Claude's responses with text and tool_use blocks:
interface AssistantRecord extends BaseRecord {
type: "assistant";
message: {
role: "assistant";
content: ContentBlock[];
usage?: {
input_tokens: number;
output_tokens: number;
cache_creation_input_tokens?: number;
cache_read_input_tokens?: number;
};
};
}
type ContentBlock =
| { type: "text"; text: string }
| { type: "tool_use"; id: string; name: string; input: Record<string, any> }
| { type: "thinking"; thinking: string };
User Records
Contain user messages or tool results (when isMeta: true):
interface UserRecord extends BaseRecord {
type: "user";
message: {
role: "user";
content: string | ToolResultBlock[];
};
isMeta?: boolean; // true = internal tool result, false = real user message
isCompactSummary?: boolean;
}
interface ToolResultBlock {
type: "tool_result";
tool_use_id: string; // links back to tool_use block's id
content: string | ContentPart[];
is_error?: boolean;
}
System Records
Session lifecycle events:
interface SystemRecord extends BaseRecord {
type: "system";
subtype?: "init" | "compact_boundary";
compactMetadata?: {
trigger: "auto" | "manual";
preTokens: number;
};
}
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.
- 4d ago First seen · 140 lines · 32 tokens per session scan A 0219bc5154f3
jsonl-parser is a skill published in the GitHub repository nyktora/noctrace (5 stars, last pushed 4mo ago), licensed MIT. It adds 32 tokens to every session and 1,102 once invoked, about $0.0002 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 skills, from other repositories
agent-inspect
Local evidence debugger and trajectory-test toolkit for TypeScript AI agents. Use when capturing framework-faithful traces, asserting TraceContract/TraceFacts, packaging Evidence v2, or inspecting local runs over read-only MCP (gettracefacts).
agent-vitals
Introspect the user's local AI agent infrastructure in real time. Use proactively: before starting tasks, before scheduling cron/timer work, when stuck, after long tasks, or when claiming something works. Exposes tools: vitalssummary, shadowlist, shadowstale, burnoutsummary, burnoutstucksessions.
agenttrace-session-audit
Audit local AI coding-agent sessions with agenttrace for cost, tool failures, latency, anomalies, health, diffs, and CI gates.
github-ci-fix-onboarding
Onboards and troubleshoots GitHub PR CI fixing by configuring GitHub CLI authentication, a non-exposed GitHub token, a matching local checkout, and a ready coding agent before running fixgithubprci. Use for first-time setup, failed prerequisites, demos, or action-shaped requests to onboard the user onto the local…
greptimedb-release
Runbook for publishing a new GreptimeDB version (tag + GitHub release + docs release-note PR) on the upstream GreptimeTeam/greptimedb repo. Use when asked to "release" / "publish" a GreptimeDB version (e.g. v1.1.0, v1.0.3).
yandex-cloud
Read Yandex Cloud through its REST API. Applies to any question about VMs, metrics, logs, audit events, Kubernetes, managed databases, serverless, networking or any other Yandex Cloud resource. Never shell out to the yc CLI — it is not how this agent reaches Yandex Cloud and is usually not installed.