Gortex Temporal Dispatch Agent Configuration

https://github.com/zzet/gortex
Admin
This project provides instructions for an agent operating within a corporate fork of gortex,...
Tokens:3,642
Snippets:44
Trust Score:8.7
License:Apache-2.0
Update:1 month ago
Show doc for...
Tokens:
Raw
### Run Benchmark on Specific Adapters

Source: https://github.com/zzet/gortex/blob/main/bench/baselines/README.md

Run the benchmark focusing only on locally installable adapters like ripgrep and probe. Use the '-against' flag to specify a comma-separated list of adapter names.

```sh
go run ./bench/baselines -against ripgrep,probe
```

--------------------------------

### Gortex Token Savings Dashboard Output

Source: https://github.com/zzet/gortex/blob/main/docs/savings.md

Example of the text-based dashboard rendered by the CLI.

```text
Gortex Token Savings
====================
Cost avoided:   $168.69 (claude-opus-4) across 1,878 calls · 11,246,094 tokens saved

Today       ████████░░░░░░░░   50.0%  saved 9,200 / 18,400 tokens   $0.14
Last 7 days ██████████░░░░░░   62.5%  saved 60,100 / 96,200 tokens  $0.90
All time    ███████████████░   93.3%  saved 11,246,094 / 12,050,716 tokens  $168.69
```

--------------------------------

### Publishing Evaluation Results to BENCHMARK.md

Source: https://github.com/zzet/gortex/blob/main/docs/04-evaluation/run.md

Example markdown structure for reporting evaluation results in BENCHMARK.md. It includes key metrics like agent performance, prompt steering impact, and links to detailed write-ups for negative cases.

```markdown
## Agent-graded evaluation

**Last run: 2026-MM-DD** · agents: Sonnet 4.6 / GPT 5.4 /
Copilot CLI · judge: Sonnet 4.6 + GPT 5.4 (agreement 87%)

| Category | (a) gortex helped | (b) no difference | (c) gortex hurt |
|----------|------------------:|------------------:|----------------:|
| Architectural explanation |  6 | 1 | 2 |
| Refactor safety           |  7 | 2 | 0 |
| Bug localization          |  5 | 2 | 2 |
| Impact analysis           |  8 | 1 | 0 |
| Contract extraction       |  6 | 3 | 0 |
| **Total**                 | 32 | 9 | 4 |

- Default-prompt vs ablation-prompt delta:  +2 (a) / 0 (b) / -1 (c)
  — gortex prompt steering helps but isn't load-bearing.
- (c) cases written up in `eval/results/2026-MM-DD/c-cases.md`
  — every loss has a public post-mortem.
```

--------------------------------

### Run Full Benchmark

Source: https://github.com/zzet/gortex/blob/main/bench/baselines/README.md

Execute the full benchmark against the gortex repository using all supported adapters.

```sh
go run ./bench/baselines
```

--------------------------------

### Initialize MCP Client Sequence

Source: https://github.com/zzet/gortex/blob/main/docs/mcp-facade-v1.md

The normative sequence for initializing a named MCP client and verifying the tool roster and capabilities.

```text
initialize(clientInfo.name = "<non-empty-client-name>")
  -> resolve the compact / hide policy for this session
  -> tools/list returns exactly 21 static compact names and no legacy names
  -> read is present
  -> capabilities(domain="read", operation="file", detail="schema")
     describes read_file without promoting it
```

--------------------------------

### Run Smoke Test for Adapters

Source: https://github.com/zzet/gortex/blob/main/bench/baselines/README.md

Execute a smoke test to verify which adapters are available without running full benchmarks. This is useful for checking basic wiring.

```sh
go run ./bench/baselines -smoke
```

--------------------------------

### Run Benchmark with JSON Output

Source: https://github.com/zzet/gortex/blob/main/bench/baselines/README.md

Execute the benchmark and output the results in JSON format to a specified file. This is useful for downstream tooling.

```sh
go run ./bench/baselines -format json -json bench/results/baselines.json
```

--------------------------------

### Regenerate dependency list

Source: https://github.com/zzet/gortex/blob/main/THIRD_PARTY_NOTICES.md

Use this command to list all modules and their versions in the current project.

```bash
go list -m -f '{{.Path}} {{.Version}}' all
```

--------------------------------

### Gortex Savings CLI Commands

Source: https://github.com/zzet/gortex/blob/main/docs/savings.md

Common CLI commands for viewing and managing token savings metrics.

```bash
# Three-bucket dashboard with USD on top
gortex savings

# Per-tool breakdown inside each bucket
gortex savings --verbose

# Headline a single model (fuzzy match: "opus" → claude-opus-4)
gortex savings --model opus

# Bucket "Today" by UTC instead of local time
gortex savings --utc

# Machine-readable output (mirrors the dashboard structure: buckets[].per_tool, cost_avoided_usd, etc.)
gortex savings --json

# Wipe cumulative totals and the event history
gortex savings --reset

# Override pricing (JSON array of {model, usd_per_m_input})
GORTEX_MODEL_PRICING_JSON='[{"model":"mycorp","usd_per_m_input":5}]' gortex savings
```

--------------------------------

### Run Full gortex Evaluation Matrix

Source: https://github.com/zzet/gortex/blob/main/docs/04-evaluation/run.md

Execute the complete evaluation across all specified tasks, agents, modes, and prompts. Ensure prerequisites like API keys and a tagged build are met. This command initiates the core evaluation process.

```sh
git rev-parse HEAD > eval/results/$(date +%Y%m%d)/HEAD.sha
```

```sh
gortex eval run \
    --task-set docs/04-evaluation/task-set.md \
    --judge-prompt docs/04-evaluation/judge-prompt.md \
    --agents sonnet-4.6,gpt-5.4,copilot \
    --corpus . \
    --out eval/results/$(date +%Y%m%d)/ \
    --max-task-tokens 50000 \
    --max-task-seconds 300
```

```sh
python3 eval/scripts/aggregate.py \
    --workdir eval/results/$(date +%Y%m%d)/ \
    --judges sonnet-4.6,gpt-5.4 \
    --out eval/results/$(date +%Y%m%d)/summary.md
```

```sh
python3 eval/scripts/spotcheck.py \
    --workdir eval/results/$(date +%Y%m%d)/ \
    --sample 5 \
    --out eval/results/$(date +%Y%m%d)/spotcheck.md
```

```sh
$EDITOR BENCHMARK.md
```

--------------------------------

### Run Partial gortex Evaluation

Source: https://github.com/zzet/gortex/blob/main/docs/04-evaluation/run.md

Execute gortex evaluation with specific agents, task categories, or modes. Useful for iterative development and testing before a full run. Note that partial run numbers should not be published as benchmarks.

```sh
# Just Sonnet 4.6 (cheapest path)
gortex eval run --agents sonnet-4.6 --task-set ...

# Just one task category (smoke before the full run)
gortex eval run --agents sonnet-4.6 \
    --task-set docs/04-evaluation/task-set.md \
    --categories "Refactor safety"

# Just the WITH mode (compare absolute quality across agents)
gortex eval run --agents sonnet-4.6,gpt-5.4,copilot \
    --modes with
```

--------------------------------

### Configure Gortex Codex Hook Modes

Source: https://github.com/zzet/gortex/blob/main/docs/agents.md

Set the GORTEX_CODEX_HOOK_MODE environment variable to define how Gortex handles tool results and hook interactions during initialization.

```bash
# Existing behavior (default).
GORTEX_CODEX_HOOK_MODE=enrich gortex init --hooks-only

# Enforce graph-detectable fallback reads/searches.
GORTEX_CODEX_HOOK_MODE=deny gortex init --hooks-only

# Rewrite only requests whose semantics can be preserved.
GORTEX_CODEX_HOOK_MODE=rewrite gortex init --hooks-only

# Replace enrichable raw PostToolUse results with graph feedback.
GORTEX_CODEX_HOOK_MODE=suppress gortex init --hooks-only
```

--------------------------------

### NDCG@10 Calculation Formula

Source: https://github.com/zzet/gortex/blob/main/bench/baselines/README.md

The formula used to compute Normalized Discounted Cumulative Gain at K=10. This metric is used to evaluate the ranking quality of retrieval systems.

```plaintext
DCG@10  = Σ (gain_i / log2(i + 2))   for i in [0, min(K, |hits|))
IDCG@10 = Σ (1     / log2(i + 2))    for i in [0, min(K, |expected|))
NDCG@10 = DCG@10 / IDCG@10
```
Tokens:
Raw